ACS Publications. Most Trusted. Most Cited. Most Read
Analytical Investigation of the Impact of Jet Geometry on Aeration Effectiveness Using Soft Computing Techniques
More

OR SEARCH CITATIONS

My Activity
Publications

Figure 1Loading Img
Download Hi-Res ImageDownload to MS-PowerPointCite This:ACS Omega 2023, 8, 35, 31811-31825
  • Open Access
Article

Analytical Investigation of the Impact of Jet Geometry on Aeration Effectiveness Using Soft Computing Techniques
Click to copy article linkArticle link copied!

  • Diksha Puri
    Diksha Puri
    School of Environmental Science, Shoolini University, Solan, Himachal Pradesh 173229, India
    More by Diksha Puri
  • Raj Kumar*
    Raj Kumar
    Department of Mechanical Engineering, Gachon University, Seongnam 13120, South Korea
    *Email: [email protected]
    More by Raj Kumar
    Orcidhttps://orcid.org/0000-0002-6507-7231
  • Parveen Sihag
    Parveen Sihag
    Department of Civil Engineering, Chandigarh University, Mohali, Punjab 140301, India
    More by Parveen Sihag
  • Mohindra Singh Thakur
    Mohindra Singh Thakur
    Department of Civil Engineering, Shoolini University, Solan, Himachal Pradesh 173229, India
    More by Mohindra Singh Thakur
  • Kahkashan Perveen
    Kahkashan Perveen
    Department of Botany & Microbiology, College of Science, King Saud University, P.O. Box 22452, Riyadh 11495, Saudi Arabia
    More by Kahkashan Perveen
  • Faisal M. Alfaisal
    Faisal M. Alfaisal
    Department of Civil Engineering, College of Engineering, King Saud University, Riyadh 11495, Saudi Arabia
    More by Faisal M. Alfaisal
  • Daeho Lee*
    Daeho Lee
    Department of Mechanical Engineering, Gachon University, Seongnam 13120, South Korea
    *Email: [email protected]
    More by Daeho Lee
    Orcidhttps://orcid.org/0000-0002-8119-9677
Open PDF

ACS Omega

Cite this: ACS Omega 2023, 8, 35, 31811–31825
Click to copy citationCitation copied!
https://doi.org/10.1021/acsomega.3c03294
Published August 22, 2023

Copyright © 2023 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Abstract

Click to copy section linkSection link copied!
High Resolution Image
Download MS PowerPoint Slide

Jet aeration is a commonly used technique for introducing air into water during wastewater treatment. In this investigation, the efficacy of different soft computing models, namely, Random Forest, Reduced Error Pruning Tree, Artificial Neural Network (ANN), Gaussian Process, and Support Vector Machine, was examined in predicting the aeration efficiency (E20) of circular and square jet configurations in an open channel flow. A total of 126 experimental data points were utilized to develop and validate these models. To assess the models’ performance, three goodness-of-fit parameters were employed: correlation coefficient (CC), root-mean-square error (RMSE), and mean absolute error (MAE). The analysis revealed that all of the developed models exhibited predictive capabilities, with CC values surpassing 0.8. Nonetheless, when it comes to predicting E20, the ANN model outperformed other soft computing models, achieving a CC of 0.9748, MAE of 0.0164, and RMSE of 0.0211. A sensitivity analysis emphasized that the angle of inclination exerted the most significant influence on the aeration in an open channel. Furthermore, the results demonstrated that square jets delivered superior aeration compared to that of circular jets under identical operating conditions.

This publication is licensed under

CC-BY-NC-ND 4.0 .
  • cc licence
  • by licence
  • nc licence
  • nd licence
Copyright © 2023 The Authors. Published by American Chemical Society

Subjects

what are subjects

1. Introduction

Click to copy section linkSection link copied!
Jet aeration is commonly linked to the development of a large air–water contact region as well as minimum energy dissipation, making it a better aerated system, especially in small tanks. There is a general consensus that the facilitation of oxygen transportation is significantly improved when an aeration device is made to create a plunging hollow jet. (1) Aeration is defined as the process of transportation, circulation, and mixing of DO within liquid, and this step is crucial in the secondary treatment of water and wastewater, as it enhances the level of dissolved oxygen, facilitating both chemical and biological oxidation. (2,3) In addition to its many applications in the treatment of wastewater, aeration and other gas transfer activities have a specific place in the management of water quality and play vital roles in both the pollution and self-purification of fresh water. For instance, to increase gas and liquid exchange in environmental sectors, plunging liquid jets can stir up a liquid pool’s surface when they come into contact with it. Larger interfacial regions between gas and liquid can be formed by using this phenomenon. Interfacial shear is one of two complementing processes that often produce air entrainment, which acts at the intermittent region where the air and water mix, shrinks the air boundary layer, and is one of two complementing mechanisms that typically produce air entrainment. The second event, however, occurs when a descending jet collides with a liquid pool, trapping air in the area around it. (4) Air is trapped at the top of the water and squeezed off, creating a huge cavity with a cylindrical bottom. This causes a bubbly plume to flow downstream beneath the impact surface of water. (5) The establishment of a significant air–water interface in the liquid serves as the basis for aeration because it causes atmospheric oxygen to be absorbed from air transported by a high-velocity discharge jet under the free liquid surface. An impinging water jet’s plume of air bubbles interacts with water in a still pool to transfer oxygen mass in the process of oxygenation by plunging jets. (6) The surface area and retention duration of bubbles below the water’s surface affect how much oxygen mass is exchanged to liquid. Smaller bubbles are preferred for oxygen transfer since their biphasic interfacial area per unit volume of gas is inversely related to their size. Oxygen is transferred and distributed in the liquid body through convection and diffusion. Due to their air/gas entrainment properties, plunging jets are common in many circumstances and have been proven to be efficient ways to aerate industrial effluents. (7,8) There are several different aeration techniques, such as surface aeration using eddy jet mixes, mechanical agitators, gas jet aerators, turbine agitators, plunging jet aeration using surface jet aerators, and static tube mixtures. (9) Every technique aims to incorporate oxygen into the healing process as effectively as possible. (10) With up to 50% of the total energy used in a wastewater treatment process, aeration is often the most expensive step. (11) To ensure that the biological wastewater treatment process is effectively aerated, simple basic units are needed. Jet aerators provide these characteristics. (12) Also, compared to conventional systems, the oxygen exchange mechanism using a sputtering jet of water is a better way to add oxygen to a pool of water. (13) Additional uses include cell bioreactor, oxygenation, and mineral-processing flotation cells. (14−16) In another industrial process that involves casting polymers and glass, enhancing the interaction between gas and liquid aeration is crucial for the effective creation of goods. (17) Based on how much electricity is used, an aeration system’s efficiency is measured by how well it can move oxygen into liquid water.
At hydraulic structures like drop shafts, weirs with various forms, stepped channels, chutes, etc., plunging jets are also used for aeration. Such initiatives enhance water quality and oxygen transmission. It was observed in research on vertical plunging hollow jets that KLa20increased with an increase in JV and that thicker jets have higher KLa20 values in all experiments for a given jet velocity. (10) Substantial efforts were also made to discover the influence of the number of jets on KLa20 and OTE when four sets of aerators with 1, 4, 8, and 16 units were investigated. The results indicated that the KLa20 and OTE are notably greater for the multiple plunging jets in air/water systems compared to a single plunging jet with the same flow area and similar conditions. (18) Studies have also been conducted to observe the influence of the nozzle design on oxygenation by plunging jets. The study examined three types of nozzles: circular, elliptical, and rectangular, with rounded edges. Experimental findings indicated distinct flow characteristics and entrainment patterns on the surface of free water jets as well as within the submerged water jet region of the receiving tank. The rectangular nozzle with rounded edges demonstrated higher KLa20 and OTE due to water jet expansion. However, it exhibited a shallower Hp compared with the other nozzles. On the other hand, the ellipse nozzle exhibited the greatest Hp. These results suggest that the ellipse nozzle is suitable for aerating deep pools, while the rectangular nozzle with rounded ends is preferable in applications requiring a high concentration of bubbles. (19) The study has shown that the geometry/shape of a plunging jet has a major impact on the depth of penetration and the total volumetric oxygen transfer coefficient. Although rectangular plunging jets with rounded edges have a significantly higher overall volumetric oxygen transfer coefficient (1.45 times) than circular plunging jets for the same flow area and other comparable conditions, it has been observed that conventional circular plunging jets have the highest penetration depth. (20) Plunging breakers are known to be formed by the action of jets plunging which contribute to wave dissipation and sediment transportation process. (21) In a study, the data was gathered through experiments conducted to examine how the θ of hollow jet aerators affects the oxygen transfer in water. Three different θ values, namely, 30, 45, and 60°, were utilized to generate hollow jets striking the pool water surface. The experimental findings demonstrated a notable influence of the hollow jet θ on the KLa20 at elevated JV. It was observed that the SOTE of 60° hollow jet aerators exhibited better performance with low-powered jets. However, when the kinetic power increased, the 30° hollow jet aerator displayed superior SOTE characteristics. (22) In another study, the impact of several variables such as JT, JV, and JL, and the arrangement of hollow jets to aerate the system have been explored. To acquire data for the KLa20 and SOTE experiments, we employed plunging hollow jets with numerous combinations of parameters. It was found that the KLa20 increases with JT and JV. This is mainly associated with the increased jet momentum which produces high shear and large turbulence in the pool of water. (23) When it comes to oxygenation efficiency, it has been observed that thinner jets are more effective in oxygenation, achieving efficiencies of up to 4.01 kg/kWh. (24,25)
Nowadays, soft computing and multicriteria decision making techniques have generated significant interest in various fields of engineering including hydraulics and water resources. (26−33) Numerous research studies have documented the utilization of soft computing techniques in the field of aeration, highlighting their usefulness in this field. PSO, GA, and hybrid ANN algorithms with HHO models have been extensively employed for studying hydraulic formations, showcasing exceptional efficacy in simulating the scouring depth downstream of a spillway caused by ski-jump flows. (34) A study was also carried out to predict the E20 in Parshall and Venturi flumes using a range of soft computing algorithms, including RF, tree-based M5P, GMDH, and MARS. (35) Their findings demonstrated that the MARS model is the superior predictor. A single plunging jet’s oxygen transfer rate in turbulent crossflow was measured using two distinct models and the complex behavior of the two-phase air–water flow was investigated using the flow visualization technique. (36)
The authors draw the conclusion that soft computing approaches are mostly used in forecasting the aeration features of “closed systems” with nozzle jets, jet thickness, jet length, etc. based on a critical evaluation, as shown in Table 1. However, in the current study, the focus is on creating soft computing-based models to estimate E20 determined by experimental work on various plunging hollow jet aeration system geometries in an open channel. To the best of the authors’ knowledge, α, D, HR, NJ, Fr. No, and jet geometries Jcir and Jsq are not taken into account when relating the E20 characteristics of hollow jets that plunge when these factors are present.
Table 1. Literature Review of Soft Computing Models Applied to Study Jet Aeration
authorsexperimental setupinput parametersjet geometrysoft computing model(s) usedreview
Kumar et al. (37)closed systemJT, JV, JL, WD-MNLR, ANN, ANFIS, MARS, GRNNANFIS using gbellmf and ANN are useful tools for modeling the oxygen transport.
Kumar et al. (38)closed systemJT, JV, JL, WD-MNLR, SVM, GP,SVM with an RBF kernel predicts the tested KLa20 to almost within a 20% scatter and performs well.
Singh et al. (39)closed systemJD, JL,HP, JV, D, P/VCircular, square, rectangular, and rectangular with rounded edge.NN, SVM, GPSVM is the best model.
Bodana et al. (40)closed systemJL, D, HPTapered and cylindrical hollow jets.ANN, GP, MNLRGP_RBF and GP_PUK are the best models for predicting the Hp entrained by the plunging hollow jets.
Onen (41)venturi aerationHP, D, θ GEP, ANN, MNLR, MLRANN is the best model.
Deswal (42)closed systemJV, JD, NJ SVM and GPSVM is the best model.
Kumar et al. (43)closed systemθ, JT, JV, Fr. No, Re. No.conical hollow jets.SVM, MNLR, M5TreeSVM model outperforms the other models to understand the oxygenation by plunging jets.
Present Studyopen channel flowα, JN, D, HR, Fr. No.circle and square hollow jets.RF, REPTree, ANN, GP, and SVMANN is the best model.
Taking into consideration the aforementioned information, the goals of the current investigation are as follows:

  • To assess the efficacy of soft computing models, RF, REPTree, ANN, GP_PUK, and SVM_RBF in understanding the jet aeration having different jet geometries.

  • To analyze the effect of jet geometry on E20.

  • Sensitivity analysis to investigate the impact of each input variable on E20.

2. Overview of Soft Computing

Click to copy section linkSection link copied!
The following section shows the soft computing techniques employed to predict E20 for the current study. The techniques used are discussed below.

2.1. Random Forest

The RF algorithm was introduced as an ensemble learning technique. (44) This technique trains and forecasts sample data using several decision trees with an identical data distribution. (45) It utilizes the tree structure to represent these rules and solves classification and regression problems. In RF, the Bootstrap approach is data- sets to resample the original data, resulting in the construction of multiple trees, each using a different bootstrap sample. (46,47) The RF approach is widely adopted in hydrology research because of its versatility and outstanding results, particularly in limited data sets. (48,49) The selection of the forest is determined by the votes of each tree. The function of the RF tree model is depicted in Figure 4, illustrating that an RF consists of a collection of classification and regression trees. The RF approach compares favorably with ANN and SVM in terms of simplicity of use and computing speed. (50) The minimal number of model parameters makes it easier to construct, and it is resistant to overfitting and can capture nonlinearity. (51) The working of the RF model tree is illustrated in Figure 1.

Figure 1

Figure 1. Working of the RF-based model.

High Resolution Image
Download MS PowerPoint Slide

2.2. Reduced Error Pruning Tree

REPTree begins with the optimal function and then prunes errors using a variance and information gain reduction method. (52) In this method, when a decision tree’s output is enormous, the decision tree is used to streamline the modeling process using training data, and the REPTree is then employed to lessen the complexity of the tree’s structure. As a result of this method, the modeling process is intended to become simpler when large input data are used by reducing the complexity of the modeling process. As it produces an effective subtree, based upon the pruning process, researchers frequently utilize the REPTree approach. (53) Using the information gain ratio, decision trees are split and pruned using two different techniques in REPTree. (54,55) Prepruning, which happens when the tree expansion is stopped during data construction, is the initial stage. This procedure is used when there are fewer occurrences than the training percentage, which shows that this node has not been disaggregated. Postpruning is the second stage, during which the tree is improved until the training set procedure is complete. The prepruning approach yields trees more quickly than the postpruning procedure, according to a comparison of the two. But compared to prepruning, postpruning results in a more accurate tree.

2.3. Artificial Neural Network

ANN, encouraged by human neural networks, is a widely parallelized processor made up of simple processing units known as neurons which are co-operating to address a specific challenge. (56) ANN model attempts to use some “organizational” principles believed to be used in humans. ANN is primarily utilized in sectors connected to information processing because it is their primary role. Numerous ANNs are employed for engineering tasks including pattern recognition, forecasting, and data compression in addition to being utilized to mimic genuine neural networks to investigate behavior and control in animals and machines. Inputs (such as synapses) are essentially multiplied by the weights in these. Weights are the strength of interunit connections that store processing capacity. (57) The weight’s value might be zero, positive, or negative. If the weight is zero, then there is no connection between the two neurons. If the weight is negative, then the signal has been blocked or diminished. To produce the desired output, ANN weights can be adjusted by using algorithms, and this approach of altering weights is called learning. The activation of the neuron is subsequently determined by these weights, which are computed by a mathematical function. This network’s neurons in this network simply add their inputs. The input multiplied by weight will be the output of the input neurons, as they only have one input. Figure 2 depicts W0–Wn as the weight and V1–Vn as the input variables.

Figure 2

Figure 2. Schematic diagram of ANN technique.

High Resolution Image
Download MS PowerPoint Slide
It has been demonstrated that ANN-based approaches can be a useful tool for simulating a variety of engineering systems in the actual world without having to solve intricate algebraic problems. (58,59) The ability to learn and identify the nonlinear relationship among the system inputs and the system outputs makes ANN useful as a black-box model. (60) Instead of calculating intricate mathematical equations, this intelligent technique imitates how the human brain responds to various challenges. (61) ANN has the capacity to generalize since they can handle unseen data faster and simpler than conventional methods after a learning phase with a few measured information sets. As consequently, researchers and scholars in a number of engineering and industrial disciplines, such as modeling and optimization, have been interested in ANN-based methodologies. (62−65)

2.4. Gaussian Process

GP is a nonparametric probabilistic technique based on kernels. It assumes that adjusted information items must share knowledge with one another. (66) Gaussian regression is the generalization of the Gaussian distribution. In the GP regression, the covariance and mean are used to represent a matrix and vectors of a Gaussian distribution. Evaluation of generalization does not need to be performed because functional dependence and data are known in advance. Based on the input test data, the Regression models using GP may distinguish between the predicted dispersion.

2.5. Support Vector Machine

SVM was developed based on statistical learning theory. (67) The primary tenet of the SVM is optimum class separation. From a limitless number of linear classifiers, SVM selects the separable classes with the lowest generalization error, or it establishes a maximum for the error produced by structural risk minimization. As a result, the greatest margin between the two classes may be calculated from the chosen hyperplane by adding the distances of the hyperplane from the points closest to the two classes. Additionally, the SVM approach offers certain advantages over other conventional methods, including a singular solution and the use of a set of multidimensional spaced kernel functions that includes intelligent nonlinear transformation.

2.6. Details on Kernel Function

How to choose a better kernel function is also a study topic because there are several kernel functions in GP and SVM. There are two common kernel functions, nevertheless, that are used for general purposes, which are represented as eqs 1 and 2, respectively.
(1)

RBF

e−γ|xi−xj|2
(1)

(2)

PUK

1[1+(||xi−xj||2221Ωσ−1)2]Ω
(2)
In this case, Gaussian noise, γ, σ, and Ω are kernel parameters. It is generally known that a suitable choice of meta-parameters affects the generalization performance (prediction accuracy) of GP and SVM, Gaussian noise, and γ, σ, and Ω influence the prediction (regression) model complexity.

2.7. Performance Evaluation Parameters

Each model’s performance was evaluated using three statistical metrics: CC, MAE, and RMSE. To assess how tight the linear connection is, the CC value is employed, and it ranges from −1 to 1. (68) The MAE method is used to determine how well the forecast matches the outcomes. It treats each error equally. MAE is used to describe errors that are evenly distributed and have a value between 0 and ∞. (69) When normally distributed errors are described, RMSE is used. It also ranges from 0 to ∞. (70)
This may be computed by using the formulas stated in eqs 3–5.
CC=∑i=1N(Ei−F¯)(Ei−E¯)∑i=1N(Fi−F¯)2∑i=1N(Ei−E¯)2
(3)
MAE=1N∑i=1N|Ei−Fi|
(4)
RMSE=1N∑i=1N(Ei−Fi)2
(5)

3. Methodology

Click to copy section linkSection link copied!
The models used for this study were trained and validated to assess their ability to predict E20 based on seven input parameters: α, D, NJ, HR, Fr. No, Jsq, and Jcir. Software known as WEKA 3.9 was utilized to create the RF, REPTree, ANN, GP, and SVM employed in this work. For optimizing each model, certain user-defined parameters must be set. The user-defined parameters for each model in the current investigation are listed in Table 2. The experiments for obtaining data sets for E20 as output for each machine learning algorithm were conducted. The data set and experimentation procedure for the investigation are given below.
Table 2. User-Defined Parameters for Applied Soft Computing Models
modelsuser-defined parameters
RF• See d = 2
REPTree• See d = 3
ANN• Learning rate = 0.3
• Hidden Layers = 600
• Momentum = 0.2
• Training Time = 500
• Neurons = 20
GP_PUK• Ω = 4
• σ = 6
• Gaussian Noise = 1
SVM_RBF• γ = 0.008

3.1. Data Set

The research focuses on the examination of E20 of jet formation within individual acrylic sheets positioned in an open channel. Essential variables, including α, ranging from 0 to 3°, D, ranging from 3.41 to 4.75 L/s, and NJ, ranging from 1 to 64, were utilized as input parameters for analysis. Additionally, E20 predictions were provided by utilizing five soft computing strategies: RF, REPTree, ANN, GP_PUK, and SVM_RBF. Utilizing the 14 acrylic sheets, each having NJ of 1,2,4,8,16,32, and 64 (Figure 3a,b), a total data set of 126 experimental readings for E20 (eqs 6–8) were recorded. The data set is further divided into training (84 readings) and testing (42 readings) sub-data sets. The data set was categorized based on an arbitrary choice. The characteristics of both collections of data sets are listed in Table 3.
Table 3. Statistical Characteristics of Training and Testing Data Sets
 αDNJHRFr. NoJsqJcirE20
Training Data Set
mean1.5001091.10518.1430.6566.0260.5000.5000.221
median1.500453.1258.0000.5225.6200.5000.5000.227
std. dev.1.2321237.20921.3640.4372.1800.5030.5030.071
kurtosis–1.5180.0120.3680.602–1.173–2.049–2.049–0.632
skewness0.0001.1701.3060.7780.2780.0000.000–0.224
minimum0.00053.2811.0000.1732.8310.0000.0000.070
maximum3.0003840.00064.0001.5649.9631.0001.0000.356
Testing Data Set
mean1.5001219.57618.1430.6566.2800.5000.5000.226
median1.500593.7508.0000.5226.5040.5000.5000.227
std. dev.1.2401576.86521.4940.4402.5240.5060.5060.072
kurtosis–1.5381.1500.4700.5630.279–2.103–2.1030.340
skewness0.0001.5771.3300.7921.0550.0000.000–0.082
minimum0.00074.2191.0000.1733.7910.0000.0000.093
maximum3.0004750.00064.0001.56411.8481.0001.0000.370

Figure 3

Figure 3. Models of jet fabricated (a) circular jets and (b) square jets.

High Resolution Image
Download MS PowerPoint Slide

3.2. Experimental Procedure

Experiments for the present study were conducted using a tilting flume (Figure 4) with dimensions of 0.45 × 0.25 × 5 m. The channel was equipped with an aeration device, strategically positioned to allow water to pass solely through hollow jets that impinged on the screens. To initiate the experiments, the water tank was filled, and an electric motor with a power of ’2 HP’ was used to circulate the water within the channel. Sodium sulfite (Na2SO3) and a catalyst, cobalt chloride (CoCl2), were mixed in the water tank to deplete oxygen in the water sample. A water sample was then collected from upstream of the aeration device, and the initial concentration of dissolved oxygen (Cn) was determined using the azide modification. (71) Subsequently, water circulation through the imping jets of the aeration device was allowed for a specified duration (time, t = 2 min). After the designated time had elapsed, the aerated/oxygenated water sample was collected from downstream of the aeration device to measure the dissolved oxygen concentration (Cm). Throughout the experiment, the water temperature (T) was measured using a digital thermometer. The value of E20 was calculated using eqs 6–8. The methodology adopted for this study is depicted in schematic form in Figure 5.

Figure 4

Figure 4. Experimental setup

High Resolution Image
Download MS PowerPoint Slide

Figure 5

Figure 5. Flowchart of the methodology.

High Resolution Image
Download MS PowerPoint Slide
To determine E20, the following equations were used. (72)
E=Cn−CmCS*−Cn=1−CS*−CmCS*−Cn=1−1r
(6)
where Cn, Cm, CS*, and r are the oxygen level in upstream, downstream locations, saturated oxygen concentration, and oxygen aeration deficient ratio, respectively.
When all of the oxygen is transported to the water, E20 is one, whereas a zero number means that no oxygen can be transmitted. The findings from various temperature trials are normalized at 20 °C using the subsequent formula to preserve consistency in measured experiments. (73)
1−E20=(1−E)1/f
(7)
The oxygen aeration efficiency (E) at a water temperature other than 20 °C may be calculated using the following time-independent formula, (73) where E20 is the oxygen transfer efficiency at 20 °C and f is the aeration exponent.
f=1+2.1×10−2(T−20)+8.25×10−5(T−20)2
(8)

4. Results

Click to copy section linkSection link copied!
The results obtained from the analysis performed by applying various soft computing techniques are described below.

4.1. Assessment of RF- and REPTree-Based Model

REPTree predictive results were obtained from training and testing data sets. The CC, RMSE, and MAE were subsequently determined on the basis of eqs 3–5. For the RF model, the training data prediction is higher than the testing data set prediction. The results of the evaluation parameters as listed in Table 4 show that CC values for training and testing data are found to be 0.9951 and 0.956, respectively. For the REPTree model, the testing data CC value of 0.8797 shows marginally good results. It is observed that the RF model performs better than REPTree in training and testing data sets. Figures 6 and 7 also display the outcomes of the RF and REPTree model training and testing. It has been found that the RF model predicts E20 more precisely than the REPTree model. Additionally, it can be observed that in the RF model (Figure 6a,b), the points are near the line of agreement; however, in the REPTree model, the points are more dispersed (Figure 7a,b). REPTree is less aggressively competitive than RF.
Table 4. Performance Assessment of the Applied Models
approachesCCRMSEMAE
Training Data Set
RF0.99510.00760.0064
REPTree0.96420.01880.0133
ANN0.97680.02040.0154
GP_PUK0.98070.01390.0097
SVM_RBF0.97440.01590.0114
Testing Data Set
RF0.9560.02310.0187
REPTree0.87970.03470.0309
ANN0.97480.02110.0164
GP_PUK0.95690.02270.0187
SVM_RBF0.96890.01770.136

Figure 6

Figure 6. Scatter plot for the observed and predicted points of E20 using RF model (a) training and (b) testing.

High Resolution Image
Download MS PowerPoint Slide

Figure 7

Figure 7. Scatter plot for observed and predicted points of E20 using REPTree Model (a) training and (b) testing.

High Resolution Image
Download MS PowerPoint Slide

4.2. Assessment of ANN-Based Model

A multilayer perceptor layer directs the iterative process of creating an ANN model. To obtain the best predictive model, three distinct performance assessment criteria were used, as shown in Table 4. Findings from the table indicate that ANN-based models perform best among other models with CC values of 0.9748, RMSE values of 0.0211, and MAE values of 0.0164. Figure 8a,b shows the training and testing phases of ANN-based models with a scatter graph showing actual and anticipated values. Most of the points on these graphs are concentrated on the line of perfect agreement, which reflects the closest match between the characteristics of the actual and projected outcomes and denotes increased dependability. Table 2 displays the user-defined parameters used to evaluate the ANN model.

Figure 8

Figure 8. Scatter plot for observed and predicted points of E20 using ANN model (a) training and (b) testing.

High Resolution Image
Download MS PowerPoint Slide

4.3. Assessment of GP_PUK-Based Model

Developing a GP model is a trial-and-error process with setting user-defined parameters to obtain the best model. The user-defined parameters set for the current study are shown in Table 2. Table 4 shows the results (CC, RMSE, and MAE) obtained by the GP_PUK model for predicting E20 with different jet geometries in an open channel. These assessment criteria values show that model GP_PUK performs relatively better than other models (Figure 9a,b) which depicts a scatter plot for the GP_PUK model using the training and testing data sets, respectively. Observing Table 4 and Figure 9, it is found that the GP_PUK model is appropriate for estimating E20 with the current data set.

Figure 9

Figure 9. Scatter plot for observed and predicted points of E20 using GP_PUK model (a) training and (b) testing.

High Resolution Image
Download MS PowerPoint Slide

4.4. Assessment of SVM_RBF-Based Model

After a trial-and-error procedure, the best user-defined parameters were chosen based on the problem’s various characteristics. After the experiment with various training functions, the best training function for the problem was determined. The values of user-defined parameters were determined by trial and error. In the present study, the RBF has been used as the kernel function to forecast E20 provided by jets having different geometries in an open channel. The optimal quantities of these parameters are assessed for the suggested approach and presented in Table 2 since they influence the accuracy of the SVM model. For the SVM_RBF forecasting model, the majority of the points are along the slanted line (Figure 10a,b). The performance of SVM_RBF appears to be fairly acceptable according to Table 4. With the data set from the current investigation, it is the second-best model for predicting jet aeration.

Figure 10

Figure 10. Scatter plot for observed and predicted points of E20 using SVM_RBF model (a) training and (b) testing.

High Resolution Image
Download MS PowerPoint Slide

4.5. Comparison of Applied Models

In this section, E20 of plunging hollow jets having different geometries (circle and square) in an open channel is examined by implementing five machine learning techniques, i.e., RF, REPTree, ANN, GP_PUK, and SVM_RBF. Seven input parameters, α, D, NJ, HR, Fr. No, Jsq, and Jcir jet geometry, as well as E20 as an output parameter for the prediction of the performance of jet geometries to achieve higher efficiency are assessed by performance evaluation parameters as given in eqs 6–8. Table 4 shows a comparison of all of the models for both training and testing stages. Out of the models applied, ANN exhibits outperforming results in the testing stage with the highest CC value of 0.9748 and least errors (RMSE = 0.0211, and MAE = 0.0164). The performance evaluation of all of the models for both training and testing stages suggests that the ANN-based model is near to actual data during the testing stage (Figure 11). The predicted relative error by the applied models is shown in Figure 12. Table 5 shows quartile values of actual and predicted E20 to compare the predicted and actual values. The best estimated model ANN has an IQR of 0.113. Figure 13 shows a boxplot which depicts that all models can estimate E20.

Figure 11

Figure 11. Comparison graph of applied soft computing models.

High Resolution Image
Download MS PowerPoint Slide

Figure 12

Figure 12. Error values of applied soft computing models in the training and testing stages.

High Resolution Image
Download MS PowerPoint Slide

Figure 13

Figure 13. Boxplot with all applied models using the testing stage.

High Resolution Image
Download MS PowerPoint Slide
Table 5. Quartile Values Using Actual and Predicted of All Applied Models for the Testing Stage
statisticactualRFREPTreeANNGP_PUKSVM_RBF
minimum0.0930.1040.0840.0620.1030.083
maximum0.370.3360.3350.4030.3360.37
first quartile0.1870.17950.160.161750.181750.1765
mean0.2262140.2204050.2201430.2191190.222190.224333
third quartile0.275250.2660.274250.274750.269250.27525
IQR0.088250.08650.114250.1130.08750.09875

5. Comparison of the Sensitivity and t-Test Analysis

Click to copy section linkSection link copied!
According to the findings, the ANN model is the most accurate among all of the models used to forecast aeration effectiveness in an open channel flow with various jet shapes. As a result, a sensitivity analysis was done to determine the most important input parameter. For sensitivity analysis, seven input parameters are employed, as indicated in Table 6. According to the results of the sensitivity study, α is the input parameter that is the most sensitive, having a smaller CC and hence more errors. This study’s main objective is to evaluate the effectiveness of circular and square jet geometry for aeration. Table 6 also reveals that square jets are more responsive than circular geometry jets, suggesting that square jets can transfer more DO under the same experimental circumstances. Hence, achieving good E20 can be accomplished by deploying square jets in an open channel.
Table 6. Results for Comparison of Sensitivity and t-Test Analysis
sensitivity analysis with best predictive model (ANN)
variables combinationparameter eliminatedCCRMSEMAE
E20 = f (HR, α, NJ, Fr. No, D, Jsq, Jcir) 0.97480.02110.0164
E20 = f (HR, NJ, Fr. No, D, Jsq, Jcir)α0.91520.03560.0273
E20 = f (HR, α, NJ, D, Jsq, Jcir)Fr. No0.9470.03860.0316
E20 = f (α, NJ, D, Fr. No, Jsq, Jcir)HR0.96620.02410.0195
E20 = f (HR, α, NJ, Fr. No, Jsq, Jcir)D0.97120.02130.0174
E20 = f (HR, α, Fr. No, D, Jsq, Jcir)NJ0.9720.02410.0188
E20 = f (HR, α, NJ, Fr. No, D, Jcir)Jsq0.97370.02010.0157
E20 = f (HR, α, NJ, Fr. No, D, Jsq)Jcir0.97430.02020.0163
experimental t-test
variableT statP value
Fr. No28.82.50 × 10–57
α11.635.24 × 10–22
HR10.981.19 × 10–20
NJ9.431.39 × 10–16
D9.391.75 × 10–16
Jsq8.211.03 × 10–13
Jcir7.828.80 × 10–13
Additionally, Table 6 also displays the t-test outcomes for the comparison of two samples, assuming variances that are not equal. The results obtained from the experimental t-test align with the sensitivity analysis acquired through the use of the modeling technique known as ANN. The input parameter α shows a high degree of sensitivity, as indicated in the sensitivity table derived from the machine learning approach. On the other hand, the parameter Fr. No exhibits the highest t-statistic value. This can be justified by acknowledging that the Fr. No is not an independent variable and does not possess direct input significance. The Fr. No is a function of the discharge rate and jet area and is calculated as:
Fr.No=Dvg∗HR
(9)
where Dv is the discharge velocity (which is 110.89 cm/s for 3.41, 124.88 cm/s for 3.84, and 154.47 cm/s for 4.75 l/s), g denotes accelerated gravity (g/cm3), and HR is the hydraulic radius for each jet.
HR=Jetareaπ×NJ×DJn2;NJ=1,2,4,8,16,32,64
(10)
where Jet area is 30.75 cm2, NJis the number of jets, and DJn is the diameter/sides of each jet geometry.
Nevertheless, it is noteworthy that in both examinations, the variables α, Fr. No, and HR consistently exhibited an identical level of significance. Moreover, upon contrasting the two jet configurations, it became evident that square jets outperformed circular jets, displaying a substantially superior performance.

6. Discussion

Click to copy section linkSection link copied!
As opposed to classical computing, soft computing uses approximation models to address complicated real-world issues. In a sense, soft computing takes its cues from the human mind. Examples of soft computing methods include fuzzy logic, GA, ANN, machine learning, and expert systems. (74) Prediction of E20 is a highly prioritized study, especially for water management resources that are heavily polluted. In order to estimate the Jsq and Jcir aeration in an open channel flow, this study explores and tests the effectiveness of the soft computing models: RF, REPTree, ANN, GP_PUK, and SVM_RBF. For model development and validation, 126 data points were obtained experimentally. The developed models’ performance is evaluated using three different goodness-of-fit parameters: CC, RMSE, and MAE. Seven different input parameter angles of α, D, HR, NJ, Fr. No, and jet geometries such as Jcir and Jsq were utilized to analyze E20. Various statistical criteria have been utilized to compare the efficiency of models. It has been observed that RF was the best-performing model with a CC value of 0.9951 during the training stage but drops to 0.956 during the calibration of model. When comparing the performances of models exhibited during the testing stage, it is observed that ANN outperforms other models with the highest CC value of 0.9748 and comparatively low RMSE of 0.0211 and MAE of 0.0164. The REPTree exhibits considerable inaccuracy during testing which is the least performing model for the current data set.
Certain studies have documented the effectiveness of employing soft computing approaches in the aeration field. ANFIS and LS-SVM were effectively applied in research related to the air entrainment rate and E20 derived from downward falling jets passing over triangular weirs. The jets’ input variables comprised discharge, drop height, and weir crest angle. The performance of these modeling methods was evaluated by comparing them to predictive equations derived from both multiple linear and multiple nonlinear regression. (75) In another study, the predictive capabilities of computational techniques (ANN, SVM, and GP) in relation to estimating total KLa20 were examined. Various configurations of plunging jets were investigated, including circular, square, rectangular, and rounded rectangle designs. (76) In another study on modeling of infiltration rate using soft computing techniques such as GEP, GRNN, and ANN. It was found that ANN outperforms other models. (77) Subsequently, three types of hybrid intelligent systems were created: an imperialist competitive algorithm (ICA)- ANN, a GA-ANN, and a PSO-ANN. These systems were specifically designed to estimate the tensile strength of rocks. The findings indicated that while all of the established predictive models demonstrate satisfactory accuracy in predicting the strength, the PSO-ANN predictive model exhibited superior performance. Consequently, it was deduced that, owing to the enhanced capability of the PSO-ANN model, it could be proposed as a novel approach for forecasting the tensile strength of rocks. (78) In another study, a pair of soft computing techniques, specifically the ANN and the Co-active Neuro-Fuzzy Inference System (CANFIS), was utilized alongside MLR to formulate a model for estimating the daily pan evaporation (Ep). Results indicated that the ANN model, incorporating all input variables and a single hidden layer, proved to be the most effective in simulating Ep. (79) In a study to compare the performance of ANFIS and ANN to predict the cement-based compressive strength, it was found that ANN was the best-performing model rather than ANFIS. (80)

7. Conclusions

Click to copy section linkSection link copied!
The investigation aimed to predict how different jet geometries affect aeration in an open system. Jsq and Jcir were used for this purpose. Several soft computing techniques, namely, RF, REPTree, ANN, GP_PUK, and SVM_RBF, were employed. The results showed that all of these methods displayed strong predictive abilities, surpassing a CC of 0.8. Among them, ANN proved to be the most appropriate method for predicting E20 based on the provided data set, achieving a CC value of 0.9748. Additionally, the predicted values for the RMSE and MAE were determined to be 0.0211 and 0.0164, respectively. Sensitivity analysis using the optimized ANN model confirmed that the tilt angle of the flume bed had a significant impact on E20. Furthermore, the findings indicated that Jsq geometry outperformed Jcir in terms of E20, a result further supported by the t-test and sensitivity analysis.

Author Information

Click to copy section linkSection link copied!
  • Corresponding Authors
  • Authors
    • Diksha Puri - School of Environmental Science, Shoolini University, Solan, Himachal Pradesh 173229, India
    • Parveen Sihag - Department of Civil Engineering, Chandigarh University, Mohali, Punjab 140301, India
    • Mohindra Singh Thakur - Department of Civil Engineering, Shoolini University, Solan, Himachal Pradesh 173229, India
    • Kahkashan Perveen - Department of Botany & Microbiology, College of Science, King Saud University, P.O. Box 22452, Riyadh 11495, Saudi Arabia
    • Faisal M. Alfaisal - Department of Civil Engineering, College of Engineering, King Saud University, Riyadh 11495, Saudi Arabia
  • Author Contributions

    D.P.: Writing and preparing the manuscript. R.K.: Writing and preparing the manuscript, methodology. P.S.: Reviewing, editing. M.S.T.: Reviewing, editing, supervision. K.P.: Reviewing, editing, supervision. F.M.A.: Reviewing, editing, supervision D.L.: Reviewing, editing, revision, supervision.

  • Notes
    The authors declare no competing financial interest.

    Statements of ethics The information has been properly cited and has been collected from published works that are in the public domain.

Acknowledgments

Click to copy section linkSection link copied!

The authors acknowledge the support provided by Researchers Supporting Project Number RSP2023R297, King Saud University, Riyadh, Saudi Arabia.

Abbreviations

Click to copy section linkSection link copied!
ANFIS

adaptive neuro-fuzzy inference system

ANNs

artificial neural networks

CC

correlation coefficient

D

discharge

DO

dissolved oxygen

E20

aeration efficiency

Fr. No

Froude number

GA

genetic algorithm

GEP

gene expression programing

GMDH

group method of data handling

GP

Gaussian process

GRNN

generalized neural network

HHO

Harris Hawk’s optimization

Hp

bubble penetration depth

HR

hydraulic radius

IQR

interquartile range

Jcir

circular jets

JD

jet diameter

JL

jet length

Jsq

square jets

JT

jet thickness

JV

jet velocity

KLa20

volumetric oxygen transfer coefficient

LS-SVM

least square support vector machine

MAE

mean absolute error

MARS

multivariate adaptive regression splines

MLR

multiple linear regression

MNLR

multiple nonlinear regression

NN

neural network

NJ

number of jets

OTE

oxygen transfer efficiency

PSO

particle swarm optimization

P/V

jet power per unit volume

PUK

Pearson VII function kernel

RBF

radial basis kernel

Re. No

Reynolds number

REPTree

reduced error pruning tree

RF

random forest

RMSE

root-mean-square error

SOTE

standard oxygen transfer efficiency

Std. Dev.

standard deviation

SVM

support vector machine

α

angle of inclination of channel bed

γ

γ

θ

jet plunging angle

σ

sigma

Ω

omega

References

Click to copy section linkSection link copied!

This article references 80 other publications.

  1. 1
    Kumar, M.; Tiwari, N. K.; Ranjan, S. Experimental study on oxygen mass transfer characteristics by plunging hollow jets. Arabian J. Sci. Eng. 2021, 46, 4521– 4532,  DOI: 10.1007/s13369-020-04975-9
    Google Scholar
    1
    Experimental Study on Oxygen Mass Transfer Characteristics by Plunging Hollow Jets
    Kumar, Munish; Tiwari, N. K.; Ranjan, Subodh
    Arabian Journal for Science and Engineering (2021), 46 (5), 4521-4532CODEN: AJSEBW; ISSN:2191-4281. (Springer)
    Aeration by plunging water jets is an effective and economical method of transferring oxygen into the receiving pool of water. A dispersion of air bubbles occurs underneath the pool surface as a high-velocity liq. jet plunges on the water surface after being ejected from a particular fall height. The present study aims to investigate the impact of jet velocity, jet length, water depth and jet thickness of plunging hollow jets in oxygenating the water in the aeration tank. Empirical correlations are proposed for the detn. of volumetric oxygen transfer coeff. based on jet velocities and jet kinetic powers. Moreover, the oxygen transfer capacity of single hollow jet, double hollow jet and a dual jet (combined hollow and solid) is tested by keeping a nearly const. jet flow area through the jet openings. The observations recorded from the exptl. anal. lead to the development of predictive relationships for estg. the values of volumetric oxygen transfer coeff. in terms of basic jet variables.
  2. 2
    Skouteris, G.; Rodriguez-Garcia, G.; Reinecke, S. F.; Hampel, U. The use of pure oxygen for aeration in aerobic wastewater treatment: A review of its potential and limitations. Bioresour. Technol. 2020, 312, 123595  DOI: 10.1016/j.biortech.2020.123595
    Google Scholar
    2
    The use of pure oxygen for aeration in aerobic wastewater treatment: A review of its potential and limitations
    Skouteris, G.; Rodriguez-Garcia, G.; Reinecke, S. F.; Hampel, U.
    Bioresource Technology (2020), 312 (), 123595CODEN: BIRTEB; ISSN:0960-8524. (Elsevier Ltd.)
    A review. In aerobic wastewater treatment, aeration is the most crit. element of the treatment system. It supplies microorganisms with the required dissolved oxygen, maintains solids in suspension and, in membrane bioreactors, it controls fouling. However, conventional activated sludge is limited to the treatment of low strength wastewaters, as higher loadings require both higher biomass and higher dissolved oxygen concns. By replacing air with pure oxygen, oxygen transfer rates increase at lower flowrates. In this work, the potential and limitations of pure oxygen aeration are reviewed. The effect of the system's operational parameters and the mixed liquor characteristics on oxygen transfer, and vice versa, are detd. Pure oxygen treats higher loadings without compromising effluent quality. Fine bubbles are more efficient in oxygen transfer due to their increased contact area. However, pure oxygen is not always essential, so it is recommended to be restricted to applications where air is not adequate.
  3. 3
    Thakre, S. B.; Bhuyar, L. B.; Deshmukh, S. J. Effect of different configurations of mechanical aerators on oxygen transfer and aeration efficiency with respect to power consumption. Int. J. Aerosp. Mech. Eng. 2008, 2 (2), 100– 108
    Google Scholar
    There is no corresponding record for this reference.
  4. 4
    Qu, X.; Goharzadeh, A.; Khezzar, L.; Molki, A. Experimental characterization of air-entrainment in a plunging jet. Exp. Therm. Fluid Sci. 2013, 44, 51– 61,  DOI: 10.1016/j.expthermflusci.2012.05.013
    Google Scholar
    There is no corresponding record for this reference.
  5. 5
    Qu, X. L.; Khezzar, L.; Danciu, D.; Labois, M.; Lakehal, D. Characterization of plunging liquid jets: A combined experimental and numerical investigation. Int. J. Multiphase Flow 2011, 37 (7), 722– 731,  DOI: 10.1016/j.ijmultiphaseflow.2011.02.006
    Google Scholar
    5
    Characterization of plunging liquid jets: A combined experimental and numerical investigation
    Qu, X. L.; Khezzar, L.; Danciu, D.; Labois, M.; Lakehal, D.
    International Journal of Multiphase Flow (2011), 37 (7), 722-731CODEN: IJMFBP; ISSN:0301-9322. (Elsevier Ltd.)
    This paper presents a combined exptl. and numerical study of the flow characteristics of round vertical liq. jets plunging into a cylindrical liq. bath. The main objective of the exptl. work consists in detg. the plunging jet flow patterns, entrained air bubble sizes and the effect of the jet velocity and variations of jet falling lengths on the jet penetration depth. The instability of the jet affected by the jet velocity and falling length is also probed. On the numerical side, two different approaches were used, namely the mixt. model approach and interface-tracking approach using the level-set technique with the std. two-equation turbulence model. The numerical results are contrasted with the exptl. data. Good agreements were found between expts. and the two modeling approaches on the jet penetration depth and entraining flow characteristics, with interface tracking rendering better predictions. However, visible differences are obsd. as to the jet instability, free surface deformation and subsequent air bubble entrainment, where interface tracking is seen to be more accurate. The CFD results support the notion that the jet with the higher flow rate thus more susceptible to surface instabilities, entrains more bubbles, reflecting in turn a smaller penetration depth as a result of momentum diffusion due to bubble concn. and generated fluctuations. The liq. av. velocity field and air concn. under tank water surface were compared to existing semi-anal. correlations. Noticeable differences were revealed as to the max. velocity at the jet centerline and assocd. bubble concn. The mixt. model predicts a higher velocity than the level-set and the theory at the early stage of jet penetration, due to a higher concn. of air that cannot rise to the surface and remain trapped around the jet head. The location of the max. air content and the peak value of air holdup are also predicted differently.
  6. 6
    Chipongo, K.; Khiadani, M. Oxygen transfer by multiple vertical plunging jets in tandem. J. Environ. Eng. 2017, 143 (1), 04016072  DOI: 10.1061/(ASCE)EE.1943-7870.0001145
    Google Scholar
    6
    Oxygen transfer by multiple vertical plunging jets in tandem
    Chipongo, Kudzai; Khiadani, Mehdi
    Journal of Environmental Engineering (Reston, VA, United States) (2017), 143 (1), 04016072/1-04016072/9CODEN: JOEEDU; ISSN:0733-9372. (American Society of Civil Engineers)
    Investigations were conducted of oxygen transfer by multiple jets in tandem configuration discharging into water flowing in a rectangular channel based on the volumetric oxygen transfer coeff., KLα20. Results from this study show that water depth in the receiving channel, no. of jets, fall height, and jet velocity influence oxygen transfer. Consistent with previous research, an optimum depth was obsd. corresponding to max. oxygen transfer, which deviated from estns. using the equation for hydraulic structures due to the low velocities of water in the channel. Since increasing the no. of jets increases KLα20, but not essentially std. oxygen transfer efficiency (SOTE) for the same jet power per unit vol. (P/V), the authors introduced the concept of an optimum P/V. This optimum was const. for any no. of jets investigated, and at this value, both KLα20 and SOTE increase with increasing no. of jets. Ultimately, a relationship is proposed for estg. KLα20 for multiple plunging jets in tandem with modifications for practical application to spatially varied flow in gutters.
  7. 7
    Mozaffari, M. H.; Shafiepour, E.; Mirbagheri, S. A.; Rakhshandehroo, G.; Wallace, S.; Stefanakis, A. I. Hydraulic characterization and removal of metals and nutrients in an aerated horizontal subsurface flow “racetrack” wetland treating primary-treated oil industry effluent. Water Res. 2021, 200, 117220  DOI: 10.1016/j.watres.2021.117220
    Google Scholar
    7
    Hydraulic characterization and removal of metals and nutrients in an aerated horizontal subsurface flow "racetrack" wetland treating primary-treated oil industry effluent
    Mozaffari, Mohammad-Hosein; Shafiepour, Ehsan; Mirbagheri, Seyed Ahmad; Rakhshandehroo, Gholamreza; Wallace, Scott; Stefanakis, Alexandros I.
    Water Research (2021), 200 (), 117220CODEN: WATRAG; ISSN:0043-1354. (Elsevier Ltd.)
    Constructed wetlands (CW) are an attractive technol. due to their operational simplicity and low life-cycle cost. It has been applied for refinery effluent treatment but mostly single-stage designs (e.g., vertical or horizontal flow) have been tested. However, to achieve a good treatment efficiency for industrial effluents, different treatment conditions (both aerobic and anaerobic) are needed. This means that hybrid CW systems are typically required with a resp. increased area demand. In addn., a strong aerobic environment that facilitates the formation of iron, manganese, zinc and aluminum ppts. cannot be established with passive wetland systems, while the role of these oxyhydroxide compds. in the further co-pptn. and removal of heavy metals such as copper, nickel, lead, and chromium that can simplify the overall treatment of industrial wastewaters is poorly understood in CW. Therefore, this study tests for the first time an innovative CW design that combines an artificially aerated section with a non-aerated section in a single unit applied for oil refinery wastewater treatment. Four pilot units were tested with different design (i.e., planted/unplanted, aerated/non-aerated) and operational (two different hydraulic loading rates) characteristics to est. the role of plants and artificial aeration and to identify the optimum design configuration. The pilot units received a primary refinery effluent, i.e., after passing through a dissolved air flotation unit. The first-order removal of heavy metals under aerobic conditions is evaluated, along with the removal of phenols and nutrients. High removal rates for Fe (96-98%), Mn (38-81%), Al (49-73%), and Zn (99-100%) generally as oxyhydroxide ppts. were found, while removal of Cu (61-80%), Ni (70-85%), Pb (96-99%) and Cr (60-92%) under aerobic conditions was also obsd., likely through co-pptn. Complete phenols and ammonia nitrogen removal was also found. The first-order rate coeff. (k) calcd. from the collected data demonstrates that the tested CW represents an advanced wetland design reaching higher removal rates at a smaller area demand than the common CW systems.
  8. 8
    Jagaba, A. H.; Kutty, S. R. M.; Fauzi, M. A. H. M.; Razali, M. A.; Hafiz, M. F. U. M.; Noor, A. Organic and Nutrient Removal From Pulp And Paper Industry Wastewater By Extended Aeration Activated Sludge System, IOP Conference Series: Earth and Environmental Science, IOP Publishing, 2021; p 012021.
    Google Scholar
    There is no corresponding record for this reference.
  9. 9
    Shukla, B. K.; Goel, A. Study on oxygen transfer by solid jet aerator with multiple openings. Eng. Sci. Technol., An Int. J. 2018, 21 (2), 255– 260,  DOI: 10.1016/j.jestch.2018.03.007
    Google Scholar
    There is no corresponding record for this reference.
  10. 10
    Deswal, S. Oxygenation by hollow plunging water jet. J. Inst. Eng. 2009, 7 (1), 40– 47,  DOI: 10.3126/jie.v7i1.2061
    Google Scholar
    There is no corresponding record for this reference.
  11. 11
    Drewnowski, J.; Remiszewska-Skwarek, A.; Duda, S.; Łagód, G. Aeration process in bioreactors as the main energy consumer in a wastewater treatment plant. Review of solutions and methods of process optimization. Processes 2019, 7 (5), 311,  DOI: 10.3390/pr7050311
    Google Scholar
    11
    Aeration process in bioreactors as the main energy consumer in a wastewater treatment plant. review of solutions and methods of process optimization
    Drewnowski, Jakub; Remiszewska-Skwarek, Anna; Duda, Sylwia; Lagod, Grzegorz
    Processes (2019), 7 (5), 311CODEN: PROCCO; ISSN:2227-9717. (MDPI AG)
    Due to the key role of the biol. decompn. process of org. compds. in wastewater treatment, a very important thing is appropriate aeration of activated sludge, because microorganisms have to be supplied with an appropriate amt. of oxygen. Aeration is one of the most energy-consuming processes in the conventional activated sludge systems of wastewater treatment technol. (may consume from 50% to 90% of electricity used by a plant), which makes it the most cost-generating process incurred by treatment plants. The paper presents the construction of aeration systems, their classification as well as parameters and factors that significantly affect the aeration process e.g., oxygen transfer efficiency, diffuser fouling, methods of dealing with diffuser fouling, diffuser selection. Addnl., there are briefly presented "smart control" systems in wastewater treatment and effect of application control strategy based on Supervisory Control and Data Acquisition system connected with the decrease in the energy consumption for aeration of bioreactors with activated sludge. It is noted that before the process is optimized, the system should be equipped with suitable metering devices. Only when relevant data is available, the improvements can be carried out. However, it's important, that the operator should regularly maintain good condition and high efficiency of diffusers.
  12. 12
    Radkevich, M.; Abdukodirova, M.; Shipilova, K.; Abdullaev, B. Determination of the Optimal Parameters of the Jet Aeration, IOP Conference Series: Earth and Environmental Science, IOP Publishing, 2021; p 012029.
    Google Scholar
    There is no corresponding record for this reference.
  13. 13
    Shukla, B. K.; Sharma, P. K.; Goel, A. Study on Oxygenation Performance of Solid Jet Aerator Having Circular Opening Corresponding to Variable Jet Length and Flow Area, Journal of Physics: Conference Series, IOP Publishing, 2020; p 012117.
    Google Scholar
    There is no corresponding record for this reference.
  14. 14
    Neto, P. M.; Nogueira, D. E.; Hashimura, Y.; Jung, S.; Pedras, B.; Berberan-Santos, M. N.; Rodrigues, C. A. Characterization of the Aeration and Hydrodynamics in Vertical-Wheel Bioreactors. Bioengineering 2022, 9 (8), 386,  DOI: 10.3390/bioengineering9080386
    Google Scholar
    There is no corresponding record for this reference.
  15. 15
    Li, S.; Wang, Y.; Lu, D.; Zheng, X.; Li, X. Improving separation efficiency of galena flotation using the Aerated Jet Flotation Cell. Physicochem. Probl. Miner. Process. 2020, 56, 513– 527,  DOI: 10.37190/ppmp/120108
    Google Scholar
    There is no corresponding record for this reference.
  16. 16
    Aytac, A.; Tuna, M. C. Development of a new generation flotation cell and monitoring of air bubbles. Water Practice Technol. 2023, 18 (1), 27– 39,  DOI: 10.2166/wpt.2022.170
    Google Scholar
    There is no corresponding record for this reference.
  17. 17
    Lorenceau, É.; Quéré, D.; Eggers, J. Air entrainment by a viscous jet plunging into a bath. Phys. Rev. Lett. 2004, 93 (25), 254501  DOI: 10.1103/PhysRevLett.93.254501
    Google Scholar
    17
    Air Entrainment by a Viscous Jet Plunging into a Bath
    Lorenceau, Elise; Quere, David; Eggers, Jens
    Physical Review Letters (2004), 93 (25), 254501/1-254501/4CODEN: PRLTAO; ISSN:0031-9007. (American Physical Society)
    A liq. jet plunging into a container of liq. often entrains a thin film of air with it, producing bubbles. This bubble prodn. is detrimental to many industrial processes, such as filling a container with a molten glass or polymer, or in coating processes. Conversely, in making a foam, one uses this effect; hence it is important to control the rate of bubble prodn. Here, we measure the amt. of air entrained by a viscous jet over a wide range of parameters and explain the phenomenon theor. Simple scaling arguments are shown to predict entrainment rates over 4 orders of magnitude in the dimensionless jet speed.
  18. 18
    Deswal, S.; Verma, D. V. S. Air-water oxygen transfer with multiple plunging jets. Water Quality Res. J. 2007, 42 (4), 295– 302,  DOI: 10.2166/wqrj.2007.031
    Google Scholar
    18
    Air-water oxygen transfer with multiple plunging jets
    Deswal, Surinder; Verma, Dharam Veer Singh
    Water Quality Research Journal of Canada (2007), 42 (4), 295-302CODEN: WQRCFA; ISSN:1201-3080. (Canadian Association on Water Quality)
    Despite numerous works reporting on O transfer by plunging jets, few studies were carried out on multiple plunging jets. The volumetric O-transfer coeff. and O-transfer efficiency of multiple plunging jets in a pool of water for different configurations in terms of varying nos. of jets and jet diams. were studied exptl. This research suggests that the volumetric O-transfer coeff. and O-transfer efficiency of the multiple plunging jets for air/water systems were significantly higher than those of a single plunging jet for same flow area and other similar conditions. A relation between the volumetric O-transfer coeff. and jet parameters is also proposed. The suggested relation predicted the volumetric O-transfer coeff. for single and multiple plunging jet(s) within a scatter of ±15%.
  19. 19
    Bagatur, T.; Baylar, A.; Sekerdag, N. The effect of nozzle type on air entrainment by plunging water jets. Water Qual. Res. J. 2002, 37 (3), 599– 612,  DOI: 10.2166/wqrj.2002.040
    Google Scholar
    19
    The effect of nozzle type on air entrainment by plunging water jets
    Bagatur, Tamer; Baylar, Ahmet; Sekerdag, Nusret
    Water Quality Research Journal of Canada (2002), 37 (3), 599-612CODEN: WQRCFA; ISSN:1201-3080. (Canadian Association on Water Quality)
    In this study, for the plunging water jet aeration system using various inclined nozzle types, bubble penetration depth, air entrainment rate, water jet expansion, effect of water jet circumference at impact point, oxygen transfer coeff., and oxygen transfer efficiency which changed depending on the water jet velocity, were researched in an air-water system. Numerous studies were conducted with circular nozzles. The present study describes expts. performed with different nozzle types. Three types of nozzles were examd., i.e., those with circular, ellipse and rectangle duct with rounded ends. Exptl. results showed that water jets produced with ellipse and rectangle duct with rounded ends nozzles have very different flow characteristics, entrainment patterns on free water jet surface, and submerged water jet region within the receiving tank. Higher air entrainment rate and oxygen transfer efficiency was obsd. in the rectangle duct with rounded ends nozzle due to water jet expansion. Bubble penetration depth, however, is lower for the rectangle duct with rounded ends nozzle than for the other nozzles. The ellipse nozzle provided the highest bubble penetration depth. These results showed that it is appropriate to use ellipse nozzle in aeration of deep pool and rectangle duct with rounded ends nozzle in the applications where high bubble concn. is desirable.
  20. 20
    Singh, S.; Deswa, S.; Pa, M. Performance analysis of plunging jets having different geometries. Int. J. Environ. Sci. 2011, 1 (6), 1154– 1167
    Google Scholar
    20
    Performance analysis of plunging jets having different geometries
    Singh, Shakti; Deswal, Surinder; Pal, Mahesh
    International Journal of Environmental Sciences (2011), 1 (6), 1154-1167CODEN: IJESMZ; ISSN:0976-4402. (Integrated Publishing Association)
    Despite numerous works reporting the oxygen transfer by circular plunging jets, few studies have been carried out on plunging jets of different geometries, namely circular, square, rectangular and rectangular with rounded edge. The exptl. study on these four geometries has revealed that jet geometry/shape has significant effect on the penetration depth and overall volumetric oxygen transfer coeff. of plunging jets. It has been obsd. that conventional circular plunging jets have highest penetration depth, but the overall volumetric oxygen transfer coeff. of rectangular with rounded edge plunging jets has been significantly higher (1.45 times) than circular plunging jet for same flow area and other similar conditions. This is due to the optimal and balanced utilization of incipient kinetic jet power by rectangular with rounded edge geometry, and hence suggests their distinct advantage over other geometries. Empirical relationships have also been proposed to est. the penetration depth and overall volumetric oxygen transfer coeff. from kinetic jet power for different geometries of plunging jets. The suggested empirical relationships can be useful in deciding the depth of aeration tank, detg. the optimum geometry/configuration and comparing the performance or oxygen mass transfer rates of different geometries and configurations of plunging jets under similar flow conditions.
  21. 21
    Chanson, H.; Aoki, S. I.; Maruyama, M. Unsteady air bubble entrainment and detrainment at a plunging breaker: dominant time scales and similarity of water level variations. Coastal Eng. 2002, 46 (2), 139– 157,  DOI: 10.1016/S0378-3839(02)00069-8
    Google Scholar
    There is no corresponding record for this reference.
  22. 22
    Kumar, M.; Ranjan, S.; Tiwari, N. K. Oxygen transfer study and modeling of plunging hollow jets. Appl. Water Sci. 2018, 8, 121,  DOI: 10.1007/s13201-018-0740-8
    Google Scholar
    There is no corresponding record for this reference.
  23. 23
    Mini, K. M.; Kaima, M. G.; Pillai, N. N. Study of Plunging Hollow-jet Aerators using Non-dimensional Parameters. J. Inst Eng. (India)-EN 2010, 91 (1), 20– 26
    Google Scholar
    There is no corresponding record for this reference.
  24. 24
    Ranjan, S. Hydraulics of jet aerators. J. Inst. Eng. (India): Environ. Eng. Div. 2008, 88, 29– 32
    Google Scholar
    There is no corresponding record for this reference.
  25. 25
    Leung, S. M.; Little, J. C.; Holst, T.; Love, N. G. Air/water oxygen transfer in a biological aerated filter. J. Environ. Eng. 2006, 132 (2), 181– 189,  DOI: 10.1061/(ASCE)0733-9372(2006)132:2(181)
    Google Scholar
    25
    Air/water oxygen transfer in a biological aerated filter
    Leung, Susanna M.; Little, John. C.; Holst, Troy; Love, Nancy G.
    Journal of Environmental Engineering (Reston, VA, United States) (2006), 132 (2), 181-189CODEN: JOEEDU; ISSN:0733-9372. (American Society of Civil Engineers)
    The O-transfer characteristics of an upflow biol. aerated filter filled with angular clay media were detd. over a wide range of gas and liq. flow rates. Liq.-side, O-transfer coeffs. (KLa) were measured using a N gas stripping method under abiotic conditions and were found to increase as both gas and liq. superficial velocity increases, with values 12-110/h based on empty bed vol. The effect of gas and liq. velocity, wastewater to clean water ratio, and temp. dependence was correlated to within ±20% of the exptl. KLa. Stagnant gas holdup is roughly double in wastewater compared to clean water, but the dynamic gas holdup is the same. The O-transfer coeff. is directly proportional to the dynamic gas holdup. Stagnant gas holdup does not affect the rate of O transfer. The results suggest that dynamic gas holdup largely dets. the specific interfacial area, whereas the interstitial liq. velocity largely controls the O-transfer coeff. (KL).
  26. 26
    Nguyen, P. T.; Ha, D. H.; Avand, M.; Jaafari, A.; Nguyen, H. D.; Al-Ansari, N.; Pham, B. T. Soft computing ensemble models based on logistic regression for groundwater potential mapping. Appl. Sci. 2020, 10 (7), 2469,  DOI: 10.3390/app10072469
    Google Scholar
    26
    Soft computing ensemble models based on logistic regression for groundwater potential mapping
    Nguyen, Phong Tung; Ha, Duong Hai; Avand, Mohammadtaghi; Jaafari, Abolfazl; Nguyen, Huu Duy; Al-Ansari, Nadhir; Van Phong, Tran; Sharma, Rohit; Kumar, Raghvendra; Van Le, Hiep; Ho, Lanh Si; Prakash, Indra; Pham, Binh Thai
    Applied Sciences (2020), 10 (7), 2469CODEN: ASPCC7; ISSN:2076-3417. (MDPI AG)
    Groundwater potential maps are one of the most important tools for the management of groundwater storage resources. In this study, we proposed four ensemble soft computing models based on logistic regression (LR) combined with the dagging (DLR), bagging (BLR), random subspace (RSSLR), and cascade generalization (CGLR) ensemble techniques for groundwater potential mapping in Dak Lak Province, Vietnam. A suite of well yield data and twelve geo-environmental factors (aspect, elevation, slope, curvature, Sediment Transport Index, Topog. Wetness Index, flow direction, rainfall, river d., soil, land use, and geol.) were used for generating the training and validation datasets required for the building and validation of the models. Based on the area under the receiver operating characteristic curve (AUC) and several other validation methods (neg. predictive value, pos. predictive value, root mean square error, accuracy, sensitivity, specificity, and Kappa), it was revealed that all four ensemble learning techniques were successful in enhancing the validation performance of the base LR model. The ensemble DLR model (AUC = 0.77) was the most successful model in identifying the groundwater potential zones in the study area, followed by the RSSLR (AUC = 0.744), BLR (AUC = 0.735), CGLR (AUC = 0.715), and single LR model (AUC = 0.71), resp. The models developed in this study and the resulting potential maps can assist decision-makers in the development of effective adaptive groundwater management plans.
  27. 27
    Singh, T. Utilization of cement bypass dust in the development of sustainable automotive brake friction composite materials. Arabian J. Chem. 2021, 14, 103324  DOI: 10.1016/j.arabjc.2021.103324
    Google Scholar
    27
    Utilization of cement bypass dust in the development of sustainable automotive brake friction composite materials
    Singh, Tej
    Arabian Journal of Chemistry (2021), 14 (9), 103324CODEN: AJCRDR; ISSN:1878-5352. (Elsevier B.V.)
    This article explored the potential of cement bypass dust, a waste produced during cement manufg., as filler in automotive brake friction composites. Five different cement bypass dust particles (10-25, 88-105, 210-250, 354-400 and 600-700μm) were used to manuf. non-asbestos/non-copper type friction materials. The composite's tribol. properties were obtained from a chase friction testing machine. Maximum friction, fade, and recovery coeffs. improve, whereas friction fluctuations and wear resistance of the brake friction composites decrease with cement bypass dust particle size. The worn surface morphol. revealed that the cement bypass dust particle size played a considerable role in forming the contact plateaus and deciding the wear behavior. Multi-objective optimization based on the ratio anal. approach was utilized to det. the composite's performance ranking.
  28. 28
    Mishra, S. K.; Dahiya, S.; Gangil, B.; Ranakoti, L.; Ranakoti, L.; Gangil, B.; Singh, T.; Sharma, S.; Boonyasopon, P.; Rangappa, S. M.; Rangappa, S. M.; Siengchin, S. Mechanical, morphological, and tribological characterization of novel walnut shell reinforced polylactic acid-based bio composites and prediction based on artificial neural network. Biomass Convers. Biorefin. 2022, 1– 12,  DOI: 10.1007/s13399-022-03670-z
    Google Scholar
    There is no corresponding record for this reference.
  29. 29
    Singh, T. Optimum design based on fabricated natural fiber reinforced automotive brake friction composites using hybrid CRITIC-MEW approach. J. Mater. Res. Technol. 2021, 14, 81– 92,  DOI: 10.1016/j.jmrt.2021.06.051
    Google Scholar
    29
    Optimum design based on fabricated natural fiber reinforced automotive brake friction composites using hybrid CRITIC-MEW approach
    Singh, Tej
    Journal of Materials Research and Technology (2021), 14 (), 81-92CODEN: JMRTAL; ISSN:2238-7854. (Elsevier B.V.)
    In this present study, the hybrid criteria importance through inter-criteria correlation (CRITIC) and multiplicative exponent weighting (MEW) optimization approach is applied to the problem of selecting an optimal brake friction formulation that satisfies max. performance requirements. Automotive brake friction composites contg. 5, 10, 15, and 20 wt. percentages of natural fibers (hemp, ramie, and pineapple) were developed. These composites analyzed for tribol. properties using a Chase testing machine following IS-2742 Part-4 std. The tribol. results, such as friction-fade (%), friction coeff., friction-recovery (%), friction fluctuations, friction-variability, friction-stability, and wear, are fixed as performance attributes to identify the most suitable friction formulation. The performance coeff. of friction (0.548) and friction-stability (0.93) remain highest for 5 wt.% pineapple fiber composites. Whereas the lowest wear (1.08 g) along with the least friction-recovery (107.54%) was exhibited by 5 wt.% hemp fiber composites. The highest friction-recovery (121.56%) corresponding to the lowest friction performance (0.501) was exhibited by 20 wt.% ramie fiber added composite. On the other hand, 5 wt.% ramie fiber added composite display lowest friction-fade (22.12%) with least friction-variability (0.330) and fluctuation (0.178). The exptl. results are found to be strongly compn.-dependent and without any pronounced trend. Consequently, it becomes difficult to prioritize the performance of formulations to choose the best among a set of composite formulations. Therefore, a multi-criteria decision-based optimization approach called CRITIC-MEW was applied; that suggests 5 wt.% ramie fiber added composite satisfies the max. preset performance criteria.
  30. 30
    Khudhair, Z. S.; Zubaidi, S. L.; Ortega-Martorell, S.; Al-Ansari, N.; Ethaib, S.; Hashim, K. A review of hybrid soft computing and data pre-processing techniques to forecast freshwater quality’s parameters: current trends and future directions. Environments 2022, 9 (7), 85,  DOI: 10.3390/environments9070085
    Google Scholar
    There is no corresponding record for this reference.
  31. 31
    Singh, T. A hybrid multiple-criteria decision-making approach for selecting optimal automotive brake friction composite. Material Design & Processing. Communications 2021, 3 (5), e266
    Google Scholar
    There is no corresponding record for this reference.
  32. 32
    Puri, D.; Sihag, P.; Thakur, M. S. A review: Aeration efficiency of hydraulic structures in diffusing DO in water. MethodsX 2023, 10, 102092  DOI: 10.1016/j.mex.2023.102092
    Google Scholar
    32
    A review: Aeration efficiency of hydraulic structures in diffusing DO in water
    Puri, Diksha; Sihag, Parveen; Thakur, M. S.
    MethodsX (2023), 10 (), 102092CODEN: METHC8; ISSN:2215-0161. (Elsevier B.V.)
    This paper contemplates the review of aeration efficiency with commonly used different aeration systems such as Venturi flumes, Weirs, Conduits, Stepped channels, In Venturi Aeration, the SAE value grows fast with the no. of air holes. In Weir Aeration, it was found that among all the different labyrinth weir structure, triangular notch weirs are known for the optimum results for air entrainment. The ANN model was developed with parameters discharge (Q) and tail water depth (Tw) which showed that Q is more influential parameter than Tw. In conduits structure, it was found that circular high head gated conduits have better aeration performance than other conduits. Aeration efficiency in Stepped channels cascades may range from 30% to 70%. The sensitivity anal. with ANN model showed that discharge (Q) followed by no. of steps (N) was the most influential parameter in E20. Bubble size was the important parameter to undertake when using bubble diffuser. The oxygen transfer efficiency (OTE) in jet diffusers was predicted developing an ANN model. It was found in sensitivity anal. that the input of 'velocity' is highly sensitive to OTE. According to literature, jets can provide OTE in the range of 1.91- 21.53kgO2/kW-hr.
  33. 33
    Singh, T.; Singh, V.; Ranakoti, L.; Kumar, S. Optimization on tribological properties of natural fiber reinforced brake friction composite materials: Effect of objective and subjective weighting methods. Polym. Test. 2023, 117, 107873  DOI: 10.1016/j.polymertesting.2022.107873
    Google Scholar
    33
    Optimization on tribological properties of natural fiber reinforced brake friction composite materials: Effect of objective and subjective weighting methods
    Singh, Tej; Singh, Vedant; Ranakoti, Lalit; Kumar, Sunil
    Polymer Testing (2023), 117 (), 107873CODEN: POTEDZ; ISSN:0142-9418. (Elsevier Ltd.)
    This research aims to study the effect of objective and subjective weighting methods in multi-attribute decision-making (MADM) and then develop a systematic framework for selecting the best natural fiber-reinforced friction composite for automobile braking applications. Therefore, sixteen friction composites with varying wt. amts. (5, 10, 15, and 20 wt%) of pineapple, ramie, hemp, and banana fibers were fabricated and evaluated for tribol. properties. The exptl. results, such as friction coeff., fade-recovery performance, friction fluctuations, wear, friction stability, and variability aspects, were discussed and considered performance attributes for selecting optimal compn. The results indicated that the incorporation of varying amts. of natural fibers has different effects on the tribol. properties, making it challenging to prioritize the performance of the composites to choose the best from the set of composite alternatives. Therefore, EDAS (evaluation based on the distance from the av. soln.) MADM approach has been applied to pick the best alternative from sixteen natural fiber-based brake friction composites. As an input to EDAS, different types of objective and subjective weighting methods were used to identify the importance of each attribute. These methods include the CRITIC (criteria importance through inter-criteria correlation), entropy, BWM (best-worst method), and AHP (analytic hierarchy process). The results show that the composite alternative with 5 wt% ramie fiber exhibits the optimal tribol. properties. The sensitivity anal. and validation reveal the robustness of the results, demonstrating that the same alternative dominates in diverse MADM and weighting conditions.
  34. 34
    Sammen, S. S.; Ghorbani, M. A.; Malik, A.; Tikhamarine, Y.; AmirRahmani, M.; Al-Ansari, N.; Chau, K. W. Enhanced artificial neural network with Harris hawks optimization for predicting scour depth downstream of ski-jump spillway. Appl. Sci. 2020, 10 (15), 5160,  DOI: 10.3390/app10155160
    Google Scholar
    34
    Enhanced artificial neural network with harris hawks optimization for predicting scour depth downstream of ski-jump spillway
    Sammen, Saad Sh.; Ghorbani, Mohammad Ali; Malik, Anurag; Tikhamarine, Yazid; AmirRahmani, Mohammad; Al-Ansari, Nadhir; Chau, Kwok-Wing
    Applied Sciences (2020), 10 (15), 5160CODEN: ASPCC7; ISSN:2076-3417. (MDPI AG)
    A spillway is a structure used to regulate the discharge flowing from hydraulic structures such as a dam. It also helps to dissipate the excess energy of water through the still basins. Therefore, it has a significant effect on the safety of the dam. One of the most serious problems that may be happening below the spillway is bed scouring, which leads to soil erosion and spillway failure. This will happen due to the high flow velocity on the spillway. In this study, an alternative to the conventional methods was employed to predict scour depth (SD) downstream of the ski-jump spillway. A novel optimization algorithm, namely, Harris hawks optimization (HHO), was proposed to enhance the performance of an artificial neural network (ANN) to predict the SD. The performance of the new hybrid ANN-HHO model was compared with two hybrid models, namely, the particle swarm optimization with ANN (ANN-PSO) model and the genetic algorithm with ANN (ANN-GA) model to illustrate the efficiency of ANN-HHO. Addnl., the results of the three hybrid models were compared with the traditional ANN and the empirical Wu model (WM) through performance metrics, viz., mean abs. error (MAE), root mean square error (RMSE), coeff. of correlation (CC), Willmott index (WI), mean abs. percentage error (MAPE), and through graphical interpretation (line, scatter, and box plots, and Taylor diagram). Results of the anal. revealed that the ANN-HHO model (MAE = 0.1760 m, RMSE = 0.2538 m) outperformed ANN-PSO (MAE = 0.2094 m, RMSE = 0.2891 m), ANN-GA (MAE = 0.2178 m, RMSE = 0.2981 m), ANN (MAE = 0.2494 m, RMSE = 0.3152 m) and WM (MAE = 0.1868 m, RMSE = 0.2701 m) models in the testing period. Besides, graphical inspection displays better accuracy of the ANN-HHO model than ANN-PSO, ANN-GA, ANN, and WM models for prediction of SD around the ski-jump spillway.
  35. 35
    Sihag, P.; Dursun, O. F.; Sammen, S. S.; Malik, A.; Chauhan, A. Prediction of aeration efficiency of parshall and modified venturi flumes: application of soft computing versus regression models. Water Supply 2021, 21 (8), 4068– 4085,  DOI: 10.2166/ws.2021.161
    Google Scholar
    There is no corresponding record for this reference.
  36. 36
    Jahromi, M. E.; Khiadani, M. Experimental study on oxygen transfer capacity of water jets discharging into turbulent cross-flow. J. Environ. Eng. 2017, 143 (6), 04017007  DOI: 10.1061/(ASCE)EE.1943-7870.0001194
    Google Scholar
    36
    Experimental study on oxygen transfer capacity of water jets discharging into turbulent cross-flow
    Jahromi, Mina Esmi; Khiadani, Mehdi
    Journal of Environmental Engineering (Reston, VA, United States) (2017), 143 (6), 04017007/1-04017007/12CODEN: JOEEDU; ISSN:0733-9372. (American Society of Civil Engineers)
    Plunging water jets are used for oxygenation purposes due to their inherent advantages such as simplicity, energy efficiency, and low operational cost. Specifically, these provide an efficient gas-liq. interfacial area for dissolving oxygen in water. Oxygen transfer by plunging jets into stagnant water has received considerable attention; however, the oxygenation capacity of plunging water jets discharging into turbulent cross-flow has received limited attention. The flow characteristics such as volumetric oxygen transfer coeff. and std. oxygen transfer efficiency are evaluated considering water jet to cross-flow velocity ratio, jet fall height, cross-flow depth, and jet impact angle. Two equations are proposed for estg. the oxygen transfer rate for a single plunging jet in a turbulent cross-flow. Moreover, the dynamic behavior of the resulting two-phase air-water flow is investigated with the aid of flow visualization.
  37. 37
    Kumar, M.; Tiwari, N. K.; Ranjan, S. Prediction of oxygen mass transfer of plunging hollow jets using regression models. ISH J. Hydraul. Eng. 2020, 26 (1), 23– 30,  DOI: 10.1080/09715010.2018.1435311
    Google Scholar
    There is no corresponding record for this reference.
  38. 38
    Kumar, M.; Tiwari, N. K.; Ranjan, S. Kernel function based regression approaches for estimating the oxygen transfer performance of plunging hollow jet aerator. J. Achiev. Mater. Manuf. Eng. 2019, 2 (2), 74– 84,  DOI: 10.5604/01.3001.0013.7917
    Google Scholar
    There is no corresponding record for this reference.
  39. 39
    Singh, S.; Deswal, S.; Pal, M. Modeling of overall volumetric oxygen transfer by plunging jets of different geometries. Int. J. Civil Struct. Eng. 2010, 1 (3), 591– 605,  DOI: 10.6088/ijcser.00202010049
    Google Scholar
    39
    Modeling of overall volumetric oxygen transfer by plunging jets of different geometries
    Singh, Shakti; Deswal, Surinder; Pal, Mahesh
    International Journal of Civil and Structural Engineering (2010), 1 (3), 591-605CODEN: IJCSKN; ISSN:0976-4399. (Integrated Publishing Association)
    The paper explores the potential of computational techniques in the modeling of overall volumetric oxygen transfer by plunging jets of different geometries, namely circular, square, rectangular and rectangular with rounded edge. The results predicted from neural network, support vector machines and Gaussian process techniques are compared in terms of correlation coeff., root mean square error and coeff. of detn. The outcome suggests the utility of all these computational techniques in the designing and performance evaluation of plunging jets oxygenation systems of different geometric shapes. However, support vector machines have performed better in comparison to other techniques. It has predicted overall volumetric oxygen transfer coeff. with a correlation coeff. of 0.985, coeff. of detn. of 0.968 and root mean square error of 0.002. Further, the scattering (within ±15%) is lowest in case of support vector machines approach. A comparison of results suggests that support vector machines works well and can be successfully used in modeling oxygen transfer by plunging jets of different geometries and configurations.
  40. 40
    Bodana, D.; Tiwari, N. M.; Ranjan, S.; Ghanekar, U. Estimation of the depth of penetration in a plunging hollow jet using artificial intelligence techniques. Arch. Mater. Sci. Eng. 2020, 2 (2), 49– 61,  DOI: 10.5604/01.3001.0014.3354
    Google Scholar
    There is no corresponding record for this reference.
  41. 41
    Onen, F. Prediction of penetration depth in a plunging water jet using soft computing approaches. Neural Computing and Applications 2014, 25, 217– 227,  DOI: 10.1007/s00521-013-1475-y
    Google Scholar
    There is no corresponding record for this reference.
  42. 42
    Deswal, S. Modeling oxygen-transfer by multiple plunging jets using support vector machines and Gaussian process regression techniques. Int. J. Civil Environ. Eng. 2011, 5 (1), 1– 6
    Google Scholar
    There is no corresponding record for this reference.
  43. 43
    Kumar, M.; Tiwari, N. K.; Ranjan, S. Soft computing based predictive modelling of oxygen transfer performance of plunging hollow jets. ISH J. Hydraul. Eng. 2022, 28 (sup1), 223– 233,  DOI: 10.1080/09715010.2020.1752831
    Google Scholar
    There is no corresponding record for this reference.
  44. 44
    Breiman, L. Random Forests; Statistics Department, University of California: Berkeley, CA, 2001; p 4720.
    Google Scholar
    There is no corresponding record for this reference.
  45. 45
    Kuhn, M.; Johnson, K.; Kuhn, M.; Johnson, K. Over-fitting and model tuning. Appl. Predict. Model. 2013, 61– 92,  DOI: 10.1007/978-1-4614-6849-3
    Google Scholar
    There is no corresponding record for this reference.
  46. 46
    Sharma, N.; Thakur, M. S.; Sihag, P.; Malik, M. A.; Kumar, R.; Abbas, M.; Saleel, C. A. Machine Learning Techniques for Evaluating Concrete Strength with Waste Marble Powder. Materials 2022, 15, 5811,  DOI: 10.3390/ma15175811
    Google Scholar
    46
    Machine Learning Techniques for Evaluating Concrete Strength with Waste Marble Powder
    Sharma, Nitisha; Thakur, Mohindra Singh; Sihag, Parveen; Malik, Mohammad Abdul; Kumar, Raj; Abbas, Mohamed; Saleel, Chanduveetil Ahamed
    Materials (2022), 15 (17), 5811CODEN: MATEG9; ISSN:1996-1944. (MDPI AG)
    The purpose of the research is to predict the compressive and flexural strengths of the concrete mix by using waste marble powder as a partial replacement of cement and sand, based on the exptl. data that was acquired from the lab. tests. In order to accomplish the goal, the models of Support vector machines, Support vector machines with bagging and Stochastic, Linear regression, and Gaussian processes were applied to the exptl. data for predicting the compressive and flexural strength of concrete. The effectiveness of models was also evaluated by using statistical criteria. Therefore, it can be inferred that the gaussian process and support vector machine methods can be used to predict the resp. outputs, i.e., flexural and compressive strength. The Gaussian process and Support vector machines Stochastic predicts better outcomes for flexural and compressive strength because it has a higher coeff. of correlation (0.8235 and 0.9462), lower mean abs. and root mean squared error values as (2.2808 and 1.8104) and (2.8527 and 2.3430), resp. Results suggest that all applied techniques are reliable for predicting the compressive and flexural strength of concrete and are able to reduce the exptl. work time. In comparison to input factors for this data set, the no. of curing days followed by the CA, C, FA, w, and MP is essential in predicting the flexural and compressive strength of a concrete mix for this data set.
  47. 47
    Tyralis, H.; Papacharalampous, G.; Langousis, A. A brief review of random forests for water scientists and practitioners and their recent history in water resources. Water 2019, 11 (5), 910,  DOI: 10.3390/w11050910
    Google Scholar
    There is no corresponding record for this reference.
  48. 48
    Zounemat-Kermani, M.; Batelaan, O.; Fadaee, M.; Hinkelmann, R. Ensemble machine learning paradigms in hydrology: A review. J. Hydrol. 2021, 598, 126266  DOI: 10.1016/j.jhydrol.2021.126266
    Google Scholar
    There is no corresponding record for this reference.
  49. 49
    Nahkala, B. A.; Kaleita, A. L.; Soupir, M. L. Empirical tool development for prairie pothole management using Ann AGNPS and random forest. Environ. Model. Soft. 2022, 147, 105241  DOI: 10.1016/j.envsoft.2021.105241
    Google Scholar
    There is no corresponding record for this reference.
  50. 50
    Mosavi, A.; Ozturk, P.; Chau, K. W. Flood prediction using machine learning models: Literature review. Water 2018, 10 (11), 1536,  DOI: 10.3390/w10111536
    Google Scholar
    There is no corresponding record for this reference.
  51. 51
    Liaw, A.; Wiener, M. Classification and regression by random Forest. R news 2002, 2 (3), 18– 22
    Google Scholar
    There is no corresponding record for this reference.
  52. 52
    Quinlan, J. R. Simplifying decision trees. Int. J. Man-Mach. Stud. 1987, 27 (3), 221– 234,  DOI: 10.1016/S0020-7373(87)80053-6
    Google Scholar
    There is no corresponding record for this reference.
  53. 53
    Mohamed, W. N. H. W.; Salleh, M. N. M.; Omar, A. H. A Comparative Study of Reduced Error Pruning Method in Decision Tree Algorithms, In 2012 IEEE International Conference on Control System, Computing and Engineering , 2012, November). 392– 397.
    Google Scholar
    There is no corresponding record for this reference.
  54. 54
    Devasena, C. (2015). Proficiency comparison of ladtree and reptree classifiers for credit risk forecast. arXiv e-prints, arXiv-1503.
    Google Scholar
    There is no corresponding record for this reference.
  55. 55
    Chen, W.; Hong, H.; Li, S.; Shahabi, H.; Wang, Y.; Wang, X.; Ahmad, B. B. Flood susceptibility modelling using novel hybrid approach of reduced-error pruning trees with bagging and random subspace ensembles. J. Hydrol. 2019, 575, 864– 873,  DOI: 10.1016/j.jhydrol.2019.05.089
    Google Scholar
    There is no corresponding record for this reference.
  56. 56
    Haykin, S. S. Neural networks and learning machines/Simon Haykin 2009,  DOI: 10.1016/j.enbuild.2021.111718 .
    Google Scholar
    There is no corresponding record for this reference.
  57. 57
    Gupta, N. Artificial neural network. Network Complex Systems 2013, 3 (1), 24– 28
    Google Scholar
    There is no corresponding record for this reference.
  58. 58
    Anitha, G. S.; Kuldeep, S. Neural Network Approach for Processing Substation Alarms. Int. J. Power Electron. Controllers Converters 2015, 1 (1), 21– 28
    Google Scholar
    There is no corresponding record for this reference.
  59. 59
    Shukla, M.; Abdelrahman, M. Artificial Neural Networks Based Steady State Security Analysis of Power Systems, Thirty-Sixth Southeastern Symposium on System Theory, 2004. Proceedings of the, IEEE, 2004; pp 266– 269.
    Google Scholar
    There is no corresponding record for this reference.
  60. 60
    Elsheikh, A. H.; Sharshir, S. W.; Abd Elaziz, M.; Kabeel, A. E.; Guilan, W.; Haiou, Z. Modeling of solar energy systems using artificial neural network: A comprehensive review. Sol. Energy 2019, 180, 622– 639,  DOI: 10.1016/j.solener.2019.01.037
    Google Scholar
    There is no corresponding record for this reference.
  61. 61
    Lu, C.; Li, S.; Lu, Z. Building energy prediction using artificial neural networks: A literature survey. Energy Build. 2022, 262, 111718  DOI: 10.1016/j.enbuild.2021.111718
    Google Scholar
    There is no corresponding record for this reference.
  62. 62
    Wang, Y.; Soutis, C.; Ando, D.; Sutou, Y.; Narita, F. Application of deep neural network learning in composites design. Eur. J. Mater. 2022, 2 (1), 118– 171,  DOI: 10.1080/26889277.2022.2053302
    Google Scholar
    There is no corresponding record for this reference.
  63. 63
    Yucel, M.; Nigdeli, S. M.; Bekdaş, G. 2020). Artificial Neural Networks (ANNs) and Solution of Civil Engineering Problems: ANNs and Prediction Applications. In Artificial Intelligence and Machine Learning Applications in Civil, Mechanical, and Industrial Engineering; IGI Global; pp 13– 38
    Google Scholar
    There is no corresponding record for this reference.
  64. 64
    Falcone, R.; Lima, C.; Martinelli, E. Soft computing techniques in structural and earthquake engineering: a literature review. Eng. Struct. 2020, 207, 110269  DOI: 10.1016/j.engstruct.2020.110269
    Google Scholar
    There is no corresponding record for this reference.
  65. 65
    Shiau, Y. H.; Yang, S. F.; Adha, R.; Muzayyanah, S. Modeling industrial energy demand in relation to subsector manufacturing output and climate change: artificial neural network insights. Sustainability 2022, 14 (5), 2896,  DOI: 10.3390/su14052896
    Google Scholar
    There is no corresponding record for this reference.
  66. 66
    Rasmussen, C. E.; Williams, C. K. Gaussian Processes for Machine Learning, MIT press: Cambridge, MA, 2006; Vol. 1, p. 159.
    Google Scholar
    There is no corresponding record for this reference.
  67. 67
    Vapnik, V. Statistical Learning Theory; John Willey & Sons. Inc.: New York, 1998.
    Google Scholar
    There is no corresponding record for this reference.
  68. 68
    Meghanathan, N. Assortativity Analysis of Real-World Network Graphs based on Centrality Metrics. Comput. Inf. Sci. 2016, 9 (3), 7– 25,  DOI: 10.5539/cis.v9n3p7
    Google Scholar
    There is no corresponding record for this reference.
  69. 69
    Schneider, P.; Xhafa, F. Anomaly Detection: Concepts and Methods. In Anomaly Detection and Complex Event Processing over IoT Data Streams , 2022; Chapter 3, pp 49– 66.
    Google Scholar
    There is no corresponding record for this reference.
  70. 70
    Willmott, C. J.; Matsuura, K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Res. 2005, 30 (1), 79– 82,  DOI: 10.3354/cr030079
    Google Scholar
    There is no corresponding record for this reference.
  71. 71
    Apha, A. WEF Standard Methods for the Examination of Waters and Wastewaters, Washington, DC , 2005.
    Google Scholar
    There is no corresponding record for this reference.
  72. 72
    Gulliver, J. S.; Wilhelms, S. C.; Parkhill, K. L. Predictive capabilities in oxygen transfer at hydraulic structures. J. Hydraul. Eng. 1998, 124 (7), 664– 671,  DOI: 10.1061/(ASCE)0733-9429(1998)124:7(664)
    Google Scholar
    There is no corresponding record for this reference.
  73. 73
    Novak, P.; Gulliver, J. S.; Wilhelms, S. C.; Parkhill, K. L. Predictive Capabilities in Oxygen Transfer at Hydraulic Structures. J. Hydraul. Eng. 1999, 125 (9), 989– 989,  DOI: 10.1061/(ASCE)0733-9429(1999)125:9(989)
    Google Scholar
    There is no corresponding record for this reference.
  74. 74
    Ibrahim, D. An overview of soft computing. Proc. Comput. Sci. 2016, 102, 34– 38,  DOI: 10.1016/j.procs.2016.09.366
    Google Scholar
    There is no corresponding record for this reference.
  75. 75
    Baylar, A.; Hanbay, D.; Ozpolat, E. An expert system for predicting aeration performance of weirs by using ANFIS. Expert Systems Appl. 2008, 35 (3), 1214– 1222,  DOI: 10.1016/j.eswa.2007.08.019
    Google Scholar
    There is no corresponding record for this reference.
  76. 76
    Singh, S.; Deswal, S.; Pal, M. Modeling of overall volumetric oxygen transfer by plunging jets of different geometries. Int. J. Civil Struct. Eng. 2010, 1 (3), 591– 605,  DOI: 10.1056/NEJMoa2023184
    Google Scholar
    76
    Modeling of overall volumetric oxygen transfer by plunging jets of different geometries
    Singh, Shakti; Deswal, Surinder; Pal, Mahesh
    International Journal of Civil and Structural Engineering (2010), 1 (3), 591-605CODEN: IJCSKN; ISSN:0976-4399. (Integrated Publishing Association)
    The paper explores the potential of computational techniques in the modeling of overall volumetric oxygen transfer by plunging jets of different geometries, namely circular, square, rectangular and rectangular with rounded edge. The results predicted from neural network, support vector machines and Gaussian process techniques are compared in terms of correlation coeff., root mean square error and coeff. of detn. The outcome suggests the utility of all these computational techniques in the designing and performance evaluation of plunging jets oxygenation systems of different geometric shapes. However, support vector machines have performed better in comparison to other techniques. It has predicted overall volumetric oxygen transfer coeff. with a correlation coeff. of 0.985, coeff. of detn. of 0.968 and root mean square error of 0.002. Further, the scattering (within ±15%) is lowest in case of support vector machines approach. A comparison of results suggests that support vector machines works well and can be successfully used in modeling oxygen transfer by plunging jets of different geometries and configurations.
  77. 77
    Sihag, P.; Singh, B.; Sepah Vand, A.; Mehdipour, V. Modeling the infiltration process with soft computing techniques. ISH J. Hydraul. Eng. 2020, 26 (2), 138– 152,  DOI: 10.1080/09715010.2018.1464408
    Google Scholar
    There is no corresponding record for this reference.
  78. 78
    Mahdiyar, A.; Armaghani, D. J.; Marto, A.; Nilashi, M.; Ismail, S. Rock tensile strength prediction using empirical and soft computing approaches. Bull. Eng. Geol. Environ. 2019, 78, 4519– 4531,  DOI: 10.1007/s10064-018-1405-4
    Google Scholar
    There is no corresponding record for this reference.
  79. 79
    Malik, A.; Kumar, A. Pan evaporation simulation based on daily meteorological data using soft computing techniques and multiple linear regression. Water Resour. Manage. 2015, 29, 1859– 1872,  DOI: 10.1007/s11269-015-0915-0
    Google Scholar
    There is no corresponding record for this reference.
  80. 80
    Armaghani, D. J.; Asteris, P. G. A comparative study of ANN and ANFIS models for the prediction of cement-based mortar materials compressive strength. Neural Comput. Appl. 2021, 33 (9), 4501– 4532,  DOI: 10.1007/s00521-020-05244-4
    Google Scholar
    There is no corresponding record for this reference.

Cited By

Click to copy section linkSection link copied!
Citation Statements
Explore this article's citation statements on scite.ai
powered by  Scite

This article is cited by 7 publications.

  1. Diksha Puri, Daeho Lee, Dhananjay Vasant khankal, Mohindra Singh Thakur, Faisal M. Alfaisal, Shamshad Alam, Raj Kumar, Mohammad Amir Khan. Decision Tree-Based Modeling of the Aeration Effectiveness of Circular Plunging Jets. ACS Omega 2023, 8 (42) , 38950-38960. https://doi.org/10.1021/acsomega.3c03375
  2. Ashwin Narendra Raut, Ahmad Alyaseen, Afzal Husain Khan, Supriya Janga, Arunava Poddar, Parveen Sihag, Vinisha Sumra, Daeho Lee, Raj Kumar, Tej Singh. RETRACTED: Incorporating crumb rubber in slag-based geopolymer: Experimental work and predictive modelling. Case Studies in Construction Materials 2024, 21 , e03893. https://doi.org/10.1016/j.cscm.2024.e03893
  3. Prachi Pramod Shedge, Pradnya Mehta, Sakshi Vyavahare, Vandana Rupnar. Enhancing Maternal Health. 2024, 65-96. https://doi.org/10.4018/979-8-3693-3711-0.ch004
  4. Vibha Yadav, Amit Kumar Yadav, Vedant Singh, Tej Singh. Artificial neural network an innovative approach in air pollutant prediction for environmental applications: A review. Results in Engineering 2024, 22 , 102305. https://doi.org/10.1016/j.rineng.2024.102305
  5. Saurabh Jaglan, Sunita Kumari, Praveen Aggarwal. Estimation of Vulnerable Road User Accident Frequency through the Soft Computing Models. Communications - Scientific letters of the University of Zilina 2024, 26 (2) , E1-E11. https://doi.org/10.26552/com.C.2024.023
  6. Saad Shamim Ansari, Ahmad Azeem, Mohammad Asad, Kamran Zafar, Syed Muhammad Ibrahim. Comparative analysis of conventional and ensemble machine learning models for predicting split tensile strength in thermal stressed SCM-blended lightweight concrete. Materials Today: Proceedings 2024, 54 https://doi.org/10.1016/j.matpr.2024.04.081
  7. Gourav Arora, Devender Kumar, Balraj Singh. Tree based Regression Models for Predicting the Compressive Strength of Concrete at High Temperature. IOP Conference Series: Earth and Environmental Science 2024, 1327 (1) , 012015. https://doi.org/10.1088/1755-1315/1327/1/012015
Download PDF
Go to ACS Omega
Get e-Alerts
Get e-Alerts

ACS Omega

Cite this: ACS Omega 2023, 8, 35, 31811–31825
Click to copy citationCitation copied!
https://doi.org/10.1021/acsomega.3c03294
Published August 22, 2023

Copyright © 2023 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Article Views

760

Altmetric

-

Citations

7
Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 1

    Figure 1. Working of the RF-based model.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 2

    Figure 2. Schematic diagram of ANN technique.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 3

    Figure 3. Models of jet fabricated (a) circular jets and (b) square jets.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 4

    Figure 4. Experimental setup

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 5

    Figure 5. Flowchart of the methodology.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 6

    Figure 6. Scatter plot for the observed and predicted points of E20 using RF model (a) training and (b) testing.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 7

    Figure 7. Scatter plot for observed and predicted points of E20 using REPTree Model (a) training and (b) testing.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 8

    Figure 8. Scatter plot for observed and predicted points of E20 using ANN model (a) training and (b) testing.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 9

    Figure 9. Scatter plot for observed and predicted points of E20 using GP_PUK model (a) training and (b) testing.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 10

    Figure 10. Scatter plot for observed and predicted points of E20 using SVM_RBF model (a) training and (b) testing.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 11

    Figure 11. Comparison graph of applied soft computing models.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 12

    Figure 12. Error values of applied soft computing models in the training and testing stages.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 13

    Figure 13. Boxplot with all applied models using the testing stage.

    High Resolution Image
    Download MS PowerPoint Slide

  • References

    This article references 80 other publications.

    1. 1
      Kumar, M.; Tiwari, N. K.; Ranjan, S. Experimental study on oxygen mass transfer characteristics by plunging hollow jets. Arabian J. Sci. Eng. 2021, 46, 4521– 4532,  DOI: 10.1007/s13369-020-04975-9
      1
      Experimental Study on Oxygen Mass Transfer Characteristics by Plunging Hollow Jets
      Kumar, Munish; Tiwari, N. K.; Ranjan, Subodh
      Arabian Journal for Science and Engineering (2021), 46 (5), 4521-4532CODEN: AJSEBW; ISSN:2191-4281. (Springer)
      Aeration by plunging water jets is an effective and economical method of transferring oxygen into the receiving pool of water. A dispersion of air bubbles occurs underneath the pool surface as a high-velocity liq. jet plunges on the water surface after being ejected from a particular fall height. The present study aims to investigate the impact of jet velocity, jet length, water depth and jet thickness of plunging hollow jets in oxygenating the water in the aeration tank. Empirical correlations are proposed for the detn. of volumetric oxygen transfer coeff. based on jet velocities and jet kinetic powers. Moreover, the oxygen transfer capacity of single hollow jet, double hollow jet and a dual jet (combined hollow and solid) is tested by keeping a nearly const. jet flow area through the jet openings. The observations recorded from the exptl. anal. lead to the development of predictive relationships for estg. the values of volumetric oxygen transfer coeff. in terms of basic jet variables.
    2. 2
      Skouteris, G.; Rodriguez-Garcia, G.; Reinecke, S. F.; Hampel, U. The use of pure oxygen for aeration in aerobic wastewater treatment: A review of its potential and limitations. Bioresour. Technol. 2020, 312, 123595  DOI: 10.1016/j.biortech.2020.123595
      2
      The use of pure oxygen for aeration in aerobic wastewater treatment: A review of its potential and limitations
      Skouteris, G.; Rodriguez-Garcia, G.; Reinecke, S. F.; Hampel, U.
      Bioresource Technology (2020), 312 (), 123595CODEN: BIRTEB; ISSN:0960-8524. (Elsevier Ltd.)
      A review. In aerobic wastewater treatment, aeration is the most crit. element of the treatment system. It supplies microorganisms with the required dissolved oxygen, maintains solids in suspension and, in membrane bioreactors, it controls fouling. However, conventional activated sludge is limited to the treatment of low strength wastewaters, as higher loadings require both higher biomass and higher dissolved oxygen concns. By replacing air with pure oxygen, oxygen transfer rates increase at lower flowrates. In this work, the potential and limitations of pure oxygen aeration are reviewed. The effect of the system's operational parameters and the mixed liquor characteristics on oxygen transfer, and vice versa, are detd. Pure oxygen treats higher loadings without compromising effluent quality. Fine bubbles are more efficient in oxygen transfer due to their increased contact area. However, pure oxygen is not always essential, so it is recommended to be restricted to applications where air is not adequate.
    3. 3
      Thakre, S. B.; Bhuyar, L. B.; Deshmukh, S. J. Effect of different configurations of mechanical aerators on oxygen transfer and aeration efficiency with respect to power consumption. Int. J. Aerosp. Mech. Eng. 2008, 2 (2), 100– 108
      There is no corresponding record for this reference.
    4. 4
      Qu, X.; Goharzadeh, A.; Khezzar, L.; Molki, A. Experimental characterization of air-entrainment in a plunging jet. Exp. Therm. Fluid Sci. 2013, 44, 51– 61,  DOI: 10.1016/j.expthermflusci.2012.05.013
      There is no corresponding record for this reference.
    5. 5
      Qu, X. L.; Khezzar, L.; Danciu, D.; Labois, M.; Lakehal, D. Characterization of plunging liquid jets: A combined experimental and numerical investigation. Int. J. Multiphase Flow 2011, 37 (7), 722– 731,  DOI: 10.1016/j.ijmultiphaseflow.2011.02.006
      5
      Characterization of plunging liquid jets: A combined experimental and numerical investigation
      Qu, X. L.; Khezzar, L.; Danciu, D.; Labois, M.; Lakehal, D.
      International Journal of Multiphase Flow (2011), 37 (7), 722-731CODEN: IJMFBP; ISSN:0301-9322. (Elsevier Ltd.)
      This paper presents a combined exptl. and numerical study of the flow characteristics of round vertical liq. jets plunging into a cylindrical liq. bath. The main objective of the exptl. work consists in detg. the plunging jet flow patterns, entrained air bubble sizes and the effect of the jet velocity and variations of jet falling lengths on the jet penetration depth. The instability of the jet affected by the jet velocity and falling length is also probed. On the numerical side, two different approaches were used, namely the mixt. model approach and interface-tracking approach using the level-set technique with the std. two-equation turbulence model. The numerical results are contrasted with the exptl. data. Good agreements were found between expts. and the two modeling approaches on the jet penetration depth and entraining flow characteristics, with interface tracking rendering better predictions. However, visible differences are obsd. as to the jet instability, free surface deformation and subsequent air bubble entrainment, where interface tracking is seen to be more accurate. The CFD results support the notion that the jet with the higher flow rate thus more susceptible to surface instabilities, entrains more bubbles, reflecting in turn a smaller penetration depth as a result of momentum diffusion due to bubble concn. and generated fluctuations. The liq. av. velocity field and air concn. under tank water surface were compared to existing semi-anal. correlations. Noticeable differences were revealed as to the max. velocity at the jet centerline and assocd. bubble concn. The mixt. model predicts a higher velocity than the level-set and the theory at the early stage of jet penetration, due to a higher concn. of air that cannot rise to the surface and remain trapped around the jet head. The location of the max. air content and the peak value of air holdup are also predicted differently.
    6. 6
      Chipongo, K.; Khiadani, M. Oxygen transfer by multiple vertical plunging jets in tandem. J. Environ. Eng. 2017, 143 (1), 04016072  DOI: 10.1061/(ASCE)EE.1943-7870.0001145
      6
      Oxygen transfer by multiple vertical plunging jets in tandem
      Chipongo, Kudzai; Khiadani, Mehdi
      Journal of Environmental Engineering (Reston, VA, United States) (2017), 143 (1), 04016072/1-04016072/9CODEN: JOEEDU; ISSN:0733-9372. (American Society of Civil Engineers)
      Investigations were conducted of oxygen transfer by multiple jets in tandem configuration discharging into water flowing in a rectangular channel based on the volumetric oxygen transfer coeff., KLα20. Results from this study show that water depth in the receiving channel, no. of jets, fall height, and jet velocity influence oxygen transfer. Consistent with previous research, an optimum depth was obsd. corresponding to max. oxygen transfer, which deviated from estns. using the equation for hydraulic structures due to the low velocities of water in the channel. Since increasing the no. of jets increases KLα20, but not essentially std. oxygen transfer efficiency (SOTE) for the same jet power per unit vol. (P/V), the authors introduced the concept of an optimum P/V. This optimum was const. for any no. of jets investigated, and at this value, both KLα20 and SOTE increase with increasing no. of jets. Ultimately, a relationship is proposed for estg. KLα20 for multiple plunging jets in tandem with modifications for practical application to spatially varied flow in gutters.
    7. 7
      Mozaffari, M. H.; Shafiepour, E.; Mirbagheri, S. A.; Rakhshandehroo, G.; Wallace, S.; Stefanakis, A. I. Hydraulic characterization and removal of metals and nutrients in an aerated horizontal subsurface flow “racetrack” wetland treating primary-treated oil industry effluent. Water Res. 2021, 200, 117220  DOI: 10.1016/j.watres.2021.117220
      7
      Hydraulic characterization and removal of metals and nutrients in an aerated horizontal subsurface flow "racetrack" wetland treating primary-treated oil industry effluent
      Mozaffari, Mohammad-Hosein; Shafiepour, Ehsan; Mirbagheri, Seyed Ahmad; Rakhshandehroo, Gholamreza; Wallace, Scott; Stefanakis, Alexandros I.
      Water Research (2021), 200 (), 117220CODEN: WATRAG; ISSN:0043-1354. (Elsevier Ltd.)
      Constructed wetlands (CW) are an attractive technol. due to their operational simplicity and low life-cycle cost. It has been applied for refinery effluent treatment but mostly single-stage designs (e.g., vertical or horizontal flow) have been tested. However, to achieve a good treatment efficiency for industrial effluents, different treatment conditions (both aerobic and anaerobic) are needed. This means that hybrid CW systems are typically required with a resp. increased area demand. In addn., a strong aerobic environment that facilitates the formation of iron, manganese, zinc and aluminum ppts. cannot be established with passive wetland systems, while the role of these oxyhydroxide compds. in the further co-pptn. and removal of heavy metals such as copper, nickel, lead, and chromium that can simplify the overall treatment of industrial wastewaters is poorly understood in CW. Therefore, this study tests for the first time an innovative CW design that combines an artificially aerated section with a non-aerated section in a single unit applied for oil refinery wastewater treatment. Four pilot units were tested with different design (i.e., planted/unplanted, aerated/non-aerated) and operational (two different hydraulic loading rates) characteristics to est. the role of plants and artificial aeration and to identify the optimum design configuration. The pilot units received a primary refinery effluent, i.e., after passing through a dissolved air flotation unit. The first-order removal of heavy metals under aerobic conditions is evaluated, along with the removal of phenols and nutrients. High removal rates for Fe (96-98%), Mn (38-81%), Al (49-73%), and Zn (99-100%) generally as oxyhydroxide ppts. were found, while removal of Cu (61-80%), Ni (70-85%), Pb (96-99%) and Cr (60-92%) under aerobic conditions was also obsd., likely through co-pptn. Complete phenols and ammonia nitrogen removal was also found. The first-order rate coeff. (k) calcd. from the collected data demonstrates that the tested CW represents an advanced wetland design reaching higher removal rates at a smaller area demand than the common CW systems.
    8. 8
      Jagaba, A. H.; Kutty, S. R. M.; Fauzi, M. A. H. M.; Razali, M. A.; Hafiz, M. F. U. M.; Noor, A. Organic and Nutrient Removal From Pulp And Paper Industry Wastewater By Extended Aeration Activated Sludge System, IOP Conference Series: Earth and Environmental Science, IOP Publishing, 2021; p 012021.
      There is no corresponding record for this reference.
    9. 9
      Shukla, B. K.; Goel, A. Study on oxygen transfer by solid jet aerator with multiple openings. Eng. Sci. Technol., An Int. J. 2018, 21 (2), 255– 260,  DOI: 10.1016/j.jestch.2018.03.007
      There is no corresponding record for this reference.
    10. 10
      Deswal, S. Oxygenation by hollow plunging water jet. J. Inst. Eng. 2009, 7 (1), 40– 47,  DOI: 10.3126/jie.v7i1.2061
      There is no corresponding record for this reference.
    11. 11
      Drewnowski, J.; Remiszewska-Skwarek, A.; Duda, S.; Łagód, G. Aeration process in bioreactors as the main energy consumer in a wastewater treatment plant. Review of solutions and methods of process optimization. Processes 2019, 7 (5), 311,  DOI: 10.3390/pr7050311
      11
      Aeration process in bioreactors as the main energy consumer in a wastewater treatment plant. review of solutions and methods of process optimization
      Drewnowski, Jakub; Remiszewska-Skwarek, Anna; Duda, Sylwia; Lagod, Grzegorz
      Processes (2019), 7 (5), 311CODEN: PROCCO; ISSN:2227-9717. (MDPI AG)
      Due to the key role of the biol. decompn. process of org. compds. in wastewater treatment, a very important thing is appropriate aeration of activated sludge, because microorganisms have to be supplied with an appropriate amt. of oxygen. Aeration is one of the most energy-consuming processes in the conventional activated sludge systems of wastewater treatment technol. (may consume from 50% to 90% of electricity used by a plant), which makes it the most cost-generating process incurred by treatment plants. The paper presents the construction of aeration systems, their classification as well as parameters and factors that significantly affect the aeration process e.g., oxygen transfer efficiency, diffuser fouling, methods of dealing with diffuser fouling, diffuser selection. Addnl., there are briefly presented "smart control" systems in wastewater treatment and effect of application control strategy based on Supervisory Control and Data Acquisition system connected with the decrease in the energy consumption for aeration of bioreactors with activated sludge. It is noted that before the process is optimized, the system should be equipped with suitable metering devices. Only when relevant data is available, the improvements can be carried out. However, it's important, that the operator should regularly maintain good condition and high efficiency of diffusers.
    12. 12
      Radkevich, M.; Abdukodirova, M.; Shipilova, K.; Abdullaev, B. Determination of the Optimal Parameters of the Jet Aeration, IOP Conference Series: Earth and Environmental Science, IOP Publishing, 2021; p 012029.
      There is no corresponding record for this reference.
    13. 13
      Shukla, B. K.; Sharma, P. K.; Goel, A. Study on Oxygenation Performance of Solid Jet Aerator Having Circular Opening Corresponding to Variable Jet Length and Flow Area, Journal of Physics: Conference Series, IOP Publishing, 2020; p 012117.
      There is no corresponding record for this reference.
    14. 14
      Neto, P. M.; Nogueira, D. E.; Hashimura, Y.; Jung, S.; Pedras, B.; Berberan-Santos, M. N.; Rodrigues, C. A. Characterization of the Aeration and Hydrodynamics in Vertical-Wheel Bioreactors. Bioengineering 2022, 9 (8), 386,  DOI: 10.3390/bioengineering9080386
      There is no corresponding record for this reference.
    15. 15
      Li, S.; Wang, Y.; Lu, D.; Zheng, X.; Li, X. Improving separation efficiency of galena flotation using the Aerated Jet Flotation Cell. Physicochem. Probl. Miner. Process. 2020, 56, 513– 527,  DOI: 10.37190/ppmp/120108
      There is no corresponding record for this reference.
    16. 16
      Aytac, A.; Tuna, M. C. Development of a new generation flotation cell and monitoring of air bubbles. Water Practice Technol. 2023, 18 (1), 27– 39,  DOI: 10.2166/wpt.2022.170
      There is no corresponding record for this reference.
    17. 17
      Lorenceau, É.; Quéré, D.; Eggers, J. Air entrainment by a viscous jet plunging into a bath. Phys. Rev. Lett. 2004, 93 (25), 254501  DOI: 10.1103/PhysRevLett.93.254501
      17
      Air Entrainment by a Viscous Jet Plunging into a Bath
      Lorenceau, Elise; Quere, David; Eggers, Jens
      Physical Review Letters (2004), 93 (25), 254501/1-254501/4CODEN: PRLTAO; ISSN:0031-9007. (American Physical Society)
      A liq. jet plunging into a container of liq. often entrains a thin film of air with it, producing bubbles. This bubble prodn. is detrimental to many industrial processes, such as filling a container with a molten glass or polymer, or in coating processes. Conversely, in making a foam, one uses this effect; hence it is important to control the rate of bubble prodn. Here, we measure the amt. of air entrained by a viscous jet over a wide range of parameters and explain the phenomenon theor. Simple scaling arguments are shown to predict entrainment rates over 4 orders of magnitude in the dimensionless jet speed.
    18. 18
      Deswal, S.; Verma, D. V. S. Air-water oxygen transfer with multiple plunging jets. Water Quality Res. J. 2007, 42 (4), 295– 302,  DOI: 10.2166/wqrj.2007.031
      18
      Air-water oxygen transfer with multiple plunging jets
      Deswal, Surinder; Verma, Dharam Veer Singh
      Water Quality Research Journal of Canada (2007), 42 (4), 295-302CODEN: WQRCFA; ISSN:1201-3080. (Canadian Association on Water Quality)
      Despite numerous works reporting on O transfer by plunging jets, few studies were carried out on multiple plunging jets. The volumetric O-transfer coeff. and O-transfer efficiency of multiple plunging jets in a pool of water for different configurations in terms of varying nos. of jets and jet diams. were studied exptl. This research suggests that the volumetric O-transfer coeff. and O-transfer efficiency of the multiple plunging jets for air/water systems were significantly higher than those of a single plunging jet for same flow area and other similar conditions. A relation between the volumetric O-transfer coeff. and jet parameters is also proposed. The suggested relation predicted the volumetric O-transfer coeff. for single and multiple plunging jet(s) within a scatter of ±15%.
    19. 19
      Bagatur, T.; Baylar, A.; Sekerdag, N. The effect of nozzle type on air entrainment by plunging water jets. Water Qual. Res. J. 2002, 37 (3), 599– 612,  DOI: 10.2166/wqrj.2002.040
      19
      The effect of nozzle type on air entrainment by plunging water jets
      Bagatur, Tamer; Baylar, Ahmet; Sekerdag, Nusret
      Water Quality Research Journal of Canada (2002), 37 (3), 599-612CODEN: WQRCFA; ISSN:1201-3080. (Canadian Association on Water Quality)
      In this study, for the plunging water jet aeration system using various inclined nozzle types, bubble penetration depth, air entrainment rate, water jet expansion, effect of water jet circumference at impact point, oxygen transfer coeff., and oxygen transfer efficiency which changed depending on the water jet velocity, were researched in an air-water system. Numerous studies were conducted with circular nozzles. The present study describes expts. performed with different nozzle types. Three types of nozzles were examd., i.e., those with circular, ellipse and rectangle duct with rounded ends. Exptl. results showed that water jets produced with ellipse and rectangle duct with rounded ends nozzles have very different flow characteristics, entrainment patterns on free water jet surface, and submerged water jet region within the receiving tank. Higher air entrainment rate and oxygen transfer efficiency was obsd. in the rectangle duct with rounded ends nozzle due to water jet expansion. Bubble penetration depth, however, is lower for the rectangle duct with rounded ends nozzle than for the other nozzles. The ellipse nozzle provided the highest bubble penetration depth. These results showed that it is appropriate to use ellipse nozzle in aeration of deep pool and rectangle duct with rounded ends nozzle in the applications where high bubble concn. is desirable.
    20. 20
      Singh, S.; Deswa, S.; Pa, M. Performance analysis of plunging jets having different geometries. Int. J. Environ. Sci. 2011, 1 (6), 1154– 1167
      20
      Performance analysis of plunging jets having different geometries
      Singh, Shakti; Deswal, Surinder; Pal, Mahesh
      International Journal of Environmental Sciences (2011), 1 (6), 1154-1167CODEN: IJESMZ; ISSN:0976-4402. (Integrated Publishing Association)
      Despite numerous works reporting the oxygen transfer by circular plunging jets, few studies have been carried out on plunging jets of different geometries, namely circular, square, rectangular and rectangular with rounded edge. The exptl. study on these four geometries has revealed that jet geometry/shape has significant effect on the penetration depth and overall volumetric oxygen transfer coeff. of plunging jets. It has been obsd. that conventional circular plunging jets have highest penetration depth, but the overall volumetric oxygen transfer coeff. of rectangular with rounded edge plunging jets has been significantly higher (1.45 times) than circular plunging jet for same flow area and other similar conditions. This is due to the optimal and balanced utilization of incipient kinetic jet power by rectangular with rounded edge geometry, and hence suggests their distinct advantage over other geometries. Empirical relationships have also been proposed to est. the penetration depth and overall volumetric oxygen transfer coeff. from kinetic jet power for different geometries of plunging jets. The suggested empirical relationships can be useful in deciding the depth of aeration tank, detg. the optimum geometry/configuration and comparing the performance or oxygen mass transfer rates of different geometries and configurations of plunging jets under similar flow conditions.
    21. 21
      Chanson, H.; Aoki, S. I.; Maruyama, M. Unsteady air bubble entrainment and detrainment at a plunging breaker: dominant time scales and similarity of water level variations. Coastal Eng. 2002, 46 (2), 139– 157,  DOI: 10.1016/S0378-3839(02)00069-8
      There is no corresponding record for this reference.
    22. 22
      Kumar, M.; Ranjan, S.; Tiwari, N. K. Oxygen transfer study and modeling of plunging hollow jets. Appl. Water Sci. 2018, 8, 121,  DOI: 10.1007/s13201-018-0740-8
      There is no corresponding record for this reference.
    23. 23
      Mini, K. M.; Kaima, M. G.; Pillai, N. N. Study of Plunging Hollow-jet Aerators using Non-dimensional Parameters. J. Inst Eng. (India)-EN 2010, 91 (1), 20– 26
      There is no corresponding record for this reference.
    24. 24
      Ranjan, S. Hydraulics of jet aerators. J. Inst. Eng. (India): Environ. Eng. Div. 2008, 88, 29– 32
      There is no corresponding record for this reference.
    25. 25
      Leung, S. M.; Little, J. C.; Holst, T.; Love, N. G. Air/water oxygen transfer in a biological aerated filter. J. Environ. Eng. 2006, 132 (2), 181– 189,  DOI: 10.1061/(ASCE)0733-9372(2006)132:2(181)
      25
      Air/water oxygen transfer in a biological aerated filter
      Leung, Susanna M.; Little, John. C.; Holst, Troy; Love, Nancy G.
      Journal of Environmental Engineering (Reston, VA, United States) (2006), 132 (2), 181-189CODEN: JOEEDU; ISSN:0733-9372. (American Society of Civil Engineers)
      The O-transfer characteristics of an upflow biol. aerated filter filled with angular clay media were detd. over a wide range of gas and liq. flow rates. Liq.-side, O-transfer coeffs. (KLa) were measured using a N gas stripping method under abiotic conditions and were found to increase as both gas and liq. superficial velocity increases, with values 12-110/h based on empty bed vol. The effect of gas and liq. velocity, wastewater to clean water ratio, and temp. dependence was correlated to within ±20% of the exptl. KLa. Stagnant gas holdup is roughly double in wastewater compared to clean water, but the dynamic gas holdup is the same. The O-transfer coeff. is directly proportional to the dynamic gas holdup. Stagnant gas holdup does not affect the rate of O transfer. The results suggest that dynamic gas holdup largely dets. the specific interfacial area, whereas the interstitial liq. velocity largely controls the O-transfer coeff. (KL).
    26. 26
      Nguyen, P. T.; Ha, D. H.; Avand, M.; Jaafari, A.; Nguyen, H. D.; Al-Ansari, N.; Pham, B. T. Soft computing ensemble models based on logistic regression for groundwater potential mapping. Appl. Sci. 2020, 10 (7), 2469,  DOI: 10.3390/app10072469
      26
      Soft computing ensemble models based on logistic regression for groundwater potential mapping
      Nguyen, Phong Tung; Ha, Duong Hai; Avand, Mohammadtaghi; Jaafari, Abolfazl; Nguyen, Huu Duy; Al-Ansari, Nadhir; Van Phong, Tran; Sharma, Rohit; Kumar, Raghvendra; Van Le, Hiep; Ho, Lanh Si; Prakash, Indra; Pham, Binh Thai
      Applied Sciences (2020), 10 (7), 2469CODEN: ASPCC7; ISSN:2076-3417. (MDPI AG)
      Groundwater potential maps are one of the most important tools for the management of groundwater storage resources. In this study, we proposed four ensemble soft computing models based on logistic regression (LR) combined with the dagging (DLR), bagging (BLR), random subspace (RSSLR), and cascade generalization (CGLR) ensemble techniques for groundwater potential mapping in Dak Lak Province, Vietnam. A suite of well yield data and twelve geo-environmental factors (aspect, elevation, slope, curvature, Sediment Transport Index, Topog. Wetness Index, flow direction, rainfall, river d., soil, land use, and geol.) were used for generating the training and validation datasets required for the building and validation of the models. Based on the area under the receiver operating characteristic curve (AUC) and several other validation methods (neg. predictive value, pos. predictive value, root mean square error, accuracy, sensitivity, specificity, and Kappa), it was revealed that all four ensemble learning techniques were successful in enhancing the validation performance of the base LR model. The ensemble DLR model (AUC = 0.77) was the most successful model in identifying the groundwater potential zones in the study area, followed by the RSSLR (AUC = 0.744), BLR (AUC = 0.735), CGLR (AUC = 0.715), and single LR model (AUC = 0.71), resp. The models developed in this study and the resulting potential maps can assist decision-makers in the development of effective adaptive groundwater management plans.
    27. 27
      Singh, T. Utilization of cement bypass dust in the development of sustainable automotive brake friction composite materials. Arabian J. Chem. 2021, 14, 103324  DOI: 10.1016/j.arabjc.2021.103324
      27
      Utilization of cement bypass dust in the development of sustainable automotive brake friction composite materials
      Singh, Tej
      Arabian Journal of Chemistry (2021), 14 (9), 103324CODEN: AJCRDR; ISSN:1878-5352. (Elsevier B.V.)
      This article explored the potential of cement bypass dust, a waste produced during cement manufg., as filler in automotive brake friction composites. Five different cement bypass dust particles (10-25, 88-105, 210-250, 354-400 and 600-700μm) were used to manuf. non-asbestos/non-copper type friction materials. The composite's tribol. properties were obtained from a chase friction testing machine. Maximum friction, fade, and recovery coeffs. improve, whereas friction fluctuations and wear resistance of the brake friction composites decrease with cement bypass dust particle size. The worn surface morphol. revealed that the cement bypass dust particle size played a considerable role in forming the contact plateaus and deciding the wear behavior. Multi-objective optimization based on the ratio anal. approach was utilized to det. the composite's performance ranking.
    28. 28
      Mishra, S. K.; Dahiya, S.; Gangil, B.; Ranakoti, L.; Ranakoti, L.; Gangil, B.; Singh, T.; Sharma, S.; Boonyasopon, P.; Rangappa, S. M.; Rangappa, S. M.; Siengchin, S. Mechanical, morphological, and tribological characterization of novel walnut shell reinforced polylactic acid-based bio composites and prediction based on artificial neural network. Biomass Convers. Biorefin. 2022, 1– 12,  DOI: 10.1007/s13399-022-03670-z
      There is no corresponding record for this reference.
    29. 29
      Singh, T. Optimum design based on fabricated natural fiber reinforced automotive brake friction composites using hybrid CRITIC-MEW approach. J. Mater. Res. Technol. 2021, 14, 81– 92,  DOI: 10.1016/j.jmrt.2021.06.051
      29
      Optimum design based on fabricated natural fiber reinforced automotive brake friction composites using hybrid CRITIC-MEW approach
      Singh, Tej
      Journal of Materials Research and Technology (2021), 14 (), 81-92CODEN: JMRTAL; ISSN:2238-7854. (Elsevier B.V.)
      In this present study, the hybrid criteria importance through inter-criteria correlation (CRITIC) and multiplicative exponent weighting (MEW) optimization approach is applied to the problem of selecting an optimal brake friction formulation that satisfies max. performance requirements. Automotive brake friction composites contg. 5, 10, 15, and 20 wt. percentages of natural fibers (hemp, ramie, and pineapple) were developed. These composites analyzed for tribol. properties using a Chase testing machine following IS-2742 Part-4 std. The tribol. results, such as friction-fade (%), friction coeff., friction-recovery (%), friction fluctuations, friction-variability, friction-stability, and wear, are fixed as performance attributes to identify the most suitable friction formulation. The performance coeff. of friction (0.548) and friction-stability (0.93) remain highest for 5 wt.% pineapple fiber composites. Whereas the lowest wear (1.08 g) along with the least friction-recovery (107.54%) was exhibited by 5 wt.% hemp fiber composites. The highest friction-recovery (121.56%) corresponding to the lowest friction performance (0.501) was exhibited by 20 wt.% ramie fiber added composite. On the other hand, 5 wt.% ramie fiber added composite display lowest friction-fade (22.12%) with least friction-variability (0.330) and fluctuation (0.178). The exptl. results are found to be strongly compn.-dependent and without any pronounced trend. Consequently, it becomes difficult to prioritize the performance of formulations to choose the best among a set of composite formulations. Therefore, a multi-criteria decision-based optimization approach called CRITIC-MEW was applied; that suggests 5 wt.% ramie fiber added composite satisfies the max. preset performance criteria.
    30. 30
      Khudhair, Z. S.; Zubaidi, S. L.; Ortega-Martorell, S.; Al-Ansari, N.; Ethaib, S.; Hashim, K. A review of hybrid soft computing and data pre-processing techniques to forecast freshwater quality’s parameters: current trends and future directions. Environments 2022, 9 (7), 85,  DOI: 10.3390/environments9070085
      There is no corresponding record for this reference.
    31. 31
      Singh, T. A hybrid multiple-criteria decision-making approach for selecting optimal automotive brake friction composite. Material Design & Processing. Communications 2021, 3 (5), e266
      There is no corresponding record for this reference.
    32. 32
      Puri, D.; Sihag, P.; Thakur, M. S. A review: Aeration efficiency of hydraulic structures in diffusing DO in water. MethodsX 2023, 10, 102092  DOI: 10.1016/j.mex.2023.102092
      32
      A review: Aeration efficiency of hydraulic structures in diffusing DO in water
      Puri, Diksha; Sihag, Parveen; Thakur, M. S.
      MethodsX (2023), 10 (), 102092CODEN: METHC8; ISSN:2215-0161. (Elsevier B.V.)
      This paper contemplates the review of aeration efficiency with commonly used different aeration systems such as Venturi flumes, Weirs, Conduits, Stepped channels, In Venturi Aeration, the SAE value grows fast with the no. of air holes. In Weir Aeration, it was found that among all the different labyrinth weir structure, triangular notch weirs are known for the optimum results for air entrainment. The ANN model was developed with parameters discharge (Q) and tail water depth (Tw) which showed that Q is more influential parameter than Tw. In conduits structure, it was found that circular high head gated conduits have better aeration performance than other conduits. Aeration efficiency in Stepped channels cascades may range from 30% to 70%. The sensitivity anal. with ANN model showed that discharge (Q) followed by no. of steps (N) was the most influential parameter in E20. Bubble size was the important parameter to undertake when using bubble diffuser. The oxygen transfer efficiency (OTE) in jet diffusers was predicted developing an ANN model. It was found in sensitivity anal. that the input of 'velocity' is highly sensitive to OTE. According to literature, jets can provide OTE in the range of 1.91- 21.53kgO2/kW-hr.
    33. 33
      Singh, T.; Singh, V.; Ranakoti, L.; Kumar, S. Optimization on tribological properties of natural fiber reinforced brake friction composite materials: Effect of objective and subjective weighting methods. Polym. Test. 2023, 117, 107873  DOI: 10.1016/j.polymertesting.2022.107873
      33
      Optimization on tribological properties of natural fiber reinforced brake friction composite materials: Effect of objective and subjective weighting methods
      Singh, Tej; Singh, Vedant; Ranakoti, Lalit; Kumar, Sunil
      Polymer Testing (2023), 117 (), 107873CODEN: POTEDZ; ISSN:0142-9418. (Elsevier Ltd.)
      This research aims to study the effect of objective and subjective weighting methods in multi-attribute decision-making (MADM) and then develop a systematic framework for selecting the best natural fiber-reinforced friction composite for automobile braking applications. Therefore, sixteen friction composites with varying wt. amts. (5, 10, 15, and 20 wt%) of pineapple, ramie, hemp, and banana fibers were fabricated and evaluated for tribol. properties. The exptl. results, such as friction coeff., fade-recovery performance, friction fluctuations, wear, friction stability, and variability aspects, were discussed and considered performance attributes for selecting optimal compn. The results indicated that the incorporation of varying amts. of natural fibers has different effects on the tribol. properties, making it challenging to prioritize the performance of the composites to choose the best from the set of composite alternatives. Therefore, EDAS (evaluation based on the distance from the av. soln.) MADM approach has been applied to pick the best alternative from sixteen natural fiber-based brake friction composites. As an input to EDAS, different types of objective and subjective weighting methods were used to identify the importance of each attribute. These methods include the CRITIC (criteria importance through inter-criteria correlation), entropy, BWM (best-worst method), and AHP (analytic hierarchy process). The results show that the composite alternative with 5 wt% ramie fiber exhibits the optimal tribol. properties. The sensitivity anal. and validation reveal the robustness of the results, demonstrating that the same alternative dominates in diverse MADM and weighting conditions.
    34. 34
      Sammen, S. S.; Ghorbani, M. A.; Malik, A.; Tikhamarine, Y.; AmirRahmani, M.; Al-Ansari, N.; Chau, K. W. Enhanced artificial neural network with Harris hawks optimization for predicting scour depth downstream of ski-jump spillway. Appl. Sci. 2020, 10 (15), 5160,  DOI: 10.3390/app10155160
      34
      Enhanced artificial neural network with harris hawks optimization for predicting scour depth downstream of ski-jump spillway
      Sammen, Saad Sh.; Ghorbani, Mohammad Ali; Malik, Anurag; Tikhamarine, Yazid; AmirRahmani, Mohammad; Al-Ansari, Nadhir; Chau, Kwok-Wing
      Applied Sciences (2020), 10 (15), 5160CODEN: ASPCC7; ISSN:2076-3417. (MDPI AG)
      A spillway is a structure used to regulate the discharge flowing from hydraulic structures such as a dam. It also helps to dissipate the excess energy of water through the still basins. Therefore, it has a significant effect on the safety of the dam. One of the most serious problems that may be happening below the spillway is bed scouring, which leads to soil erosion and spillway failure. This will happen due to the high flow velocity on the spillway. In this study, an alternative to the conventional methods was employed to predict scour depth (SD) downstream of the ski-jump spillway. A novel optimization algorithm, namely, Harris hawks optimization (HHO), was proposed to enhance the performance of an artificial neural network (ANN) to predict the SD. The performance of the new hybrid ANN-HHO model was compared with two hybrid models, namely, the particle swarm optimization with ANN (ANN-PSO) model and the genetic algorithm with ANN (ANN-GA) model to illustrate the efficiency of ANN-HHO. Addnl., the results of the three hybrid models were compared with the traditional ANN and the empirical Wu model (WM) through performance metrics, viz., mean abs. error (MAE), root mean square error (RMSE), coeff. of correlation (CC), Willmott index (WI), mean abs. percentage error (MAPE), and through graphical interpretation (line, scatter, and box plots, and Taylor diagram). Results of the anal. revealed that the ANN-HHO model (MAE = 0.1760 m, RMSE = 0.2538 m) outperformed ANN-PSO (MAE = 0.2094 m, RMSE = 0.2891 m), ANN-GA (MAE = 0.2178 m, RMSE = 0.2981 m), ANN (MAE = 0.2494 m, RMSE = 0.3152 m) and WM (MAE = 0.1868 m, RMSE = 0.2701 m) models in the testing period. Besides, graphical inspection displays better accuracy of the ANN-HHO model than ANN-PSO, ANN-GA, ANN, and WM models for prediction of SD around the ski-jump spillway.
    35. 35
      Sihag, P.; Dursun, O. F.; Sammen, S. S.; Malik, A.; Chauhan, A. Prediction of aeration efficiency of parshall and modified venturi flumes: application of soft computing versus regression models. Water Supply 2021, 21 (8), 4068– 4085,  DOI: 10.2166/ws.2021.161
      There is no corresponding record for this reference.
    36. 36
      Jahromi, M. E.; Khiadani, M. Experimental study on oxygen transfer capacity of water jets discharging into turbulent cross-flow. J. Environ. Eng. 2017, 143 (6), 04017007  DOI: 10.1061/(ASCE)EE.1943-7870.0001194
      36
      Experimental study on oxygen transfer capacity of water jets discharging into turbulent cross-flow
      Jahromi, Mina Esmi; Khiadani, Mehdi
      Journal of Environmental Engineering (Reston, VA, United States) (2017), 143 (6), 04017007/1-04017007/12CODEN: JOEEDU; ISSN:0733-9372. (American Society of Civil Engineers)
      Plunging water jets are used for oxygenation purposes due to their inherent advantages such as simplicity, energy efficiency, and low operational cost. Specifically, these provide an efficient gas-liq. interfacial area for dissolving oxygen in water. Oxygen transfer by plunging jets into stagnant water has received considerable attention; however, the oxygenation capacity of plunging water jets discharging into turbulent cross-flow has received limited attention. The flow characteristics such as volumetric oxygen transfer coeff. and std. oxygen transfer efficiency are evaluated considering water jet to cross-flow velocity ratio, jet fall height, cross-flow depth, and jet impact angle. Two equations are proposed for estg. the oxygen transfer rate for a single plunging jet in a turbulent cross-flow. Moreover, the dynamic behavior of the resulting two-phase air-water flow is investigated with the aid of flow visualization.
    37. 37
      Kumar, M.; Tiwari, N. K.; Ranjan, S. Prediction of oxygen mass transfer of plunging hollow jets using regression models. ISH J. Hydraul. Eng. 2020, 26 (1), 23– 30,  DOI: 10.1080/09715010.2018.1435311
      There is no corresponding record for this reference.
    38. 38
      Kumar, M.; Tiwari, N. K.; Ranjan, S. Kernel function based regression approaches for estimating the oxygen transfer performance of plunging hollow jet aerator. J. Achiev. Mater. Manuf. Eng. 2019, 2 (2), 74– 84,  DOI: 10.5604/01.3001.0013.7917
      There is no corresponding record for this reference.
    39. 39
      Singh, S.; Deswal, S.; Pal, M. Modeling of overall volumetric oxygen transfer by plunging jets of different geometries. Int. J. Civil Struct. Eng. 2010, 1 (3), 591– 605,  DOI: 10.6088/ijcser.00202010049
      39
      Modeling of overall volumetric oxygen transfer by plunging jets of different geometries
      Singh, Shakti; Deswal, Surinder; Pal, Mahesh
      International Journal of Civil and Structural Engineering (2010), 1 (3), 591-605CODEN: IJCSKN; ISSN:0976-4399. (Integrated Publishing Association)
      The paper explores the potential of computational techniques in the modeling of overall volumetric oxygen transfer by plunging jets of different geometries, namely circular, square, rectangular and rectangular with rounded edge. The results predicted from neural network, support vector machines and Gaussian process techniques are compared in terms of correlation coeff., root mean square error and coeff. of detn. The outcome suggests the utility of all these computational techniques in the designing and performance evaluation of plunging jets oxygenation systems of different geometric shapes. However, support vector machines have performed better in comparison to other techniques. It has predicted overall volumetric oxygen transfer coeff. with a correlation coeff. of 0.985, coeff. of detn. of 0.968 and root mean square error of 0.002. Further, the scattering (within ±15%) is lowest in case of support vector machines approach. A comparison of results suggests that support vector machines works well and can be successfully used in modeling oxygen transfer by plunging jets of different geometries and configurations.
    40. 40
      Bodana, D.; Tiwari, N. M.; Ranjan, S.; Ghanekar, U. Estimation of the depth of penetration in a plunging hollow jet using artificial intelligence techniques. Arch. Mater. Sci. Eng. 2020, 2 (2), 49– 61,  DOI: 10.5604/01.3001.0014.3354
      There is no corresponding record for this reference.
    41. 41
      Onen, F. Prediction of penetration depth in a plunging water jet using soft computing approaches. Neural Computing and Applications 2014, 25, 217– 227,  DOI: 10.1007/s00521-013-1475-y
      There is no corresponding record for this reference.
    42. 42
      Deswal, S. Modeling oxygen-transfer by multiple plunging jets using support vector machines and Gaussian process regression techniques. Int. J. Civil Environ. Eng. 2011, 5 (1), 1– 6
      There is no corresponding record for this reference.
    43. 43
      Kumar, M.; Tiwari, N. K.; Ranjan, S. Soft computing based predictive modelling of oxygen transfer performance of plunging hollow jets. ISH J. Hydraul. Eng. 2022, 28 (sup1), 223– 233,  DOI: 10.1080/09715010.2020.1752831
      There is no corresponding record for this reference.
    44. 44
      Breiman, L. Random Forests; Statistics Department, University of California: Berkeley, CA, 2001; p 4720.
      There is no corresponding record for this reference.
    45. 45
      Kuhn, M.; Johnson, K.; Kuhn, M.; Johnson, K. Over-fitting and model tuning. Appl. Predict. Model. 2013, 61– 92,  DOI: 10.1007/978-1-4614-6849-3
      There is no corresponding record for this reference.
    46. 46
      Sharma, N.; Thakur, M. S.; Sihag, P.; Malik, M. A.; Kumar, R.; Abbas, M.; Saleel, C. A. Machine Learning Techniques for Evaluating Concrete Strength with Waste Marble Powder. Materials 2022, 15, 5811,  DOI: 10.3390/ma15175811
      46
      Machine Learning Techniques for Evaluating Concrete Strength with Waste Marble Powder
      Sharma, Nitisha; Thakur, Mohindra Singh; Sihag, Parveen; Malik, Mohammad Abdul; Kumar, Raj; Abbas, Mohamed; Saleel, Chanduveetil Ahamed
      Materials (2022), 15 (17), 5811CODEN: MATEG9; ISSN:1996-1944. (MDPI AG)
      The purpose of the research is to predict the compressive and flexural strengths of the concrete mix by using waste marble powder as a partial replacement of cement and sand, based on the exptl. data that was acquired from the lab. tests. In order to accomplish the goal, the models of Support vector machines, Support vector machines with bagging and Stochastic, Linear regression, and Gaussian processes were applied to the exptl. data for predicting the compressive and flexural strength of concrete. The effectiveness of models was also evaluated by using statistical criteria. Therefore, it can be inferred that the gaussian process and support vector machine methods can be used to predict the resp. outputs, i.e., flexural and compressive strength. The Gaussian process and Support vector machines Stochastic predicts better outcomes for flexural and compressive strength because it has a higher coeff. of correlation (0.8235 and 0.9462), lower mean abs. and root mean squared error values as (2.2808 and 1.8104) and (2.8527 and 2.3430), resp. Results suggest that all applied techniques are reliable for predicting the compressive and flexural strength of concrete and are able to reduce the exptl. work time. In comparison to input factors for this data set, the no. of curing days followed by the CA, C, FA, w, and MP is essential in predicting the flexural and compressive strength of a concrete mix for this data set.
    47. 47
      Tyralis, H.; Papacharalampous, G.; Langousis, A. A brief review of random forests for water scientists and practitioners and their recent history in water resources. Water 2019, 11 (5), 910,  DOI: 10.3390/w11050910
      There is no corresponding record for this reference.
    48. 48
      Zounemat-Kermani, M.; Batelaan, O.; Fadaee, M.; Hinkelmann, R. Ensemble machine learning paradigms in hydrology: A review. J. Hydrol. 2021, 598, 126266  DOI: 10.1016/j.jhydrol.2021.126266
      There is no corresponding record for this reference.
    49. 49
      Nahkala, B. A.; Kaleita, A. L.; Soupir, M. L. Empirical tool development for prairie pothole management using Ann AGNPS and random forest. Environ. Model. Soft. 2022, 147, 105241  DOI: 10.1016/j.envsoft.2021.105241
      There is no corresponding record for this reference.
    50. 50
      Mosavi, A.; Ozturk, P.; Chau, K. W. Flood prediction using machine learning models: Literature review. Water 2018, 10 (11), 1536,  DOI: 10.3390/w10111536
      There is no corresponding record for this reference.
    51. 51
      Liaw, A.; Wiener, M. Classification and regression by random Forest. R news 2002, 2 (3), 18– 22
      There is no corresponding record for this reference.
    52. 52
      Quinlan, J. R. Simplifying decision trees. Int. J. Man-Mach. Stud. 1987, 27 (3), 221– 234,  DOI: 10.1016/S0020-7373(87)80053-6
      There is no corresponding record for this reference.
    53. 53
      Mohamed, W. N. H. W.; Salleh, M. N. M.; Omar, A. H. A Comparative Study of Reduced Error Pruning Method in Decision Tree Algorithms, In 2012 IEEE International Conference on Control System, Computing and Engineering , 2012, November). 392– 397.
      There is no corresponding record for this reference.
    54. 54
      Devasena, C. (2015). Proficiency comparison of ladtree and reptree classifiers for credit risk forecast. arXiv e-prints, arXiv-1503.
      There is no corresponding record for this reference.
    55. 55
      Chen, W.; Hong, H.; Li, S.; Shahabi, H.; Wang, Y.; Wang, X.; Ahmad, B. B. Flood susceptibility modelling using novel hybrid approach of reduced-error pruning trees with bagging and random subspace ensembles. J. Hydrol. 2019, 575, 864– 873,  DOI: 10.1016/j.jhydrol.2019.05.089
      There is no corresponding record for this reference.
    56. 56
      Haykin, S. S. Neural networks and learning machines/Simon Haykin 2009,  DOI: 10.1016/j.enbuild.2021.111718 .
      There is no corresponding record for this reference.
    57. 57
      Gupta, N. Artificial neural network. Network Complex Systems 2013, 3 (1), 24– 28
      There is no corresponding record for this reference.
    58. 58
      Anitha, G. S.; Kuldeep, S. Neural Network Approach for Processing Substation Alarms. Int. J. Power Electron. Controllers Converters 2015, 1 (1), 21– 28
      There is no corresponding record for this reference.
    59. 59
      Shukla, M.; Abdelrahman, M. Artificial Neural Networks Based Steady State Security Analysis of Power Systems, Thirty-Sixth Southeastern Symposium on System Theory, 2004. Proceedings of the, IEEE, 2004; pp 266– 269.
      There is no corresponding record for this reference.
    60. 60
      Elsheikh, A. H.; Sharshir, S. W.; Abd Elaziz, M.; Kabeel, A. E.; Guilan, W.; Haiou, Z. Modeling of solar energy systems using artificial neural network: A comprehensive review. Sol. Energy 2019, 180, 622– 639,  DOI: 10.1016/j.solener.2019.01.037
      There is no corresponding record for this reference.
    61. 61
      Lu, C.; Li, S.; Lu, Z. Building energy prediction using artificial neural networks: A literature survey. Energy Build. 2022, 262, 111718  DOI: 10.1016/j.enbuild.2021.111718
      There is no corresponding record for this reference.
    62. 62
      Wang, Y.; Soutis, C.; Ando, D.; Sutou, Y.; Narita, F. Application of deep neural network learning in composites design. Eur. J. Mater. 2022, 2 (1), 118– 171,  DOI: 10.1080/26889277.2022.2053302
      There is no corresponding record for this reference.
    63. 63
      Yucel, M.; Nigdeli, S. M.; Bekdaş, G. 2020). Artificial Neural Networks (ANNs) and Solution of Civil Engineering Problems: ANNs and Prediction Applications. In Artificial Intelligence and Machine Learning Applications in Civil, Mechanical, and Industrial Engineering; IGI Global; pp 13– 38
      There is no corresponding record for this reference.
    64. 64
      Falcone, R.; Lima, C.; Martinelli, E. Soft computing techniques in structural and earthquake engineering: a literature review. Eng. Struct. 2020, 207, 110269  DOI: 10.1016/j.engstruct.2020.110269
      There is no corresponding record for this reference.
    65. 65
      Shiau, Y. H.; Yang, S. F.; Adha, R.; Muzayyanah, S. Modeling industrial energy demand in relation to subsector manufacturing output and climate change: artificial neural network insights. Sustainability 2022, 14 (5), 2896,  DOI: 10.3390/su14052896
      There is no corresponding record for this reference.
    66. 66
      Rasmussen, C. E.; Williams, C. K. Gaussian Processes for Machine Learning, MIT press: Cambridge, MA, 2006; Vol. 1, p. 159.
      There is no corresponding record for this reference.
    67. 67
      Vapnik, V. Statistical Learning Theory; John Willey & Sons. Inc.: New York, 1998.
      There is no corresponding record for this reference.
    68. 68
      Meghanathan, N. Assortativity Analysis of Real-World Network Graphs based on Centrality Metrics. Comput. Inf. Sci. 2016, 9 (3), 7– 25,  DOI: 10.5539/cis.v9n3p7
      There is no corresponding record for this reference.
    69. 69
      Schneider, P.; Xhafa, F. Anomaly Detection: Concepts and Methods. In Anomaly Detection and Complex Event Processing over IoT Data Streams , 2022; Chapter 3, pp 49– 66.
      There is no corresponding record for this reference.
    70. 70
      Willmott, C. J.; Matsuura, K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Res. 2005, 30 (1), 79– 82,  DOI: 10.3354/cr030079
      There is no corresponding record for this reference.
    71. 71
      Apha, A. WEF Standard Methods for the Examination of Waters and Wastewaters, Washington, DC , 2005.
      There is no corresponding record for this reference.
    72. 72
      Gulliver, J. S.; Wilhelms, S. C.; Parkhill, K. L. Predictive capabilities in oxygen transfer at hydraulic structures. J. Hydraul. Eng. 1998, 124 (7), 664– 671,  DOI: 10.1061/(ASCE)0733-9429(1998)124:7(664)
      There is no corresponding record for this reference.
    73. 73
      Novak, P.; Gulliver, J. S.; Wilhelms, S. C.; Parkhill, K. L. Predictive Capabilities in Oxygen Transfer at Hydraulic Structures. J. Hydraul. Eng. 1999, 125 (9), 989– 989,  DOI: 10.1061/(ASCE)0733-9429(1999)125:9(989)
      There is no corresponding record for this reference.
    74. 74
      Ibrahim, D. An overview of soft computing. Proc. Comput. Sci. 2016, 102, 34– 38,  DOI: 10.1016/j.procs.2016.09.366
      There is no corresponding record for this reference.
    75. 75
      Baylar, A.; Hanbay, D.; Ozpolat, E. An expert system for predicting aeration performance of weirs by using ANFIS. Expert Systems Appl. 2008, 35 (3), 1214– 1222,  DOI: 10.1016/j.eswa.2007.08.019
      There is no corresponding record for this reference.
    76. 76
      Singh, S.; Deswal, S.; Pal, M. Modeling of overall volumetric oxygen transfer by plunging jets of different geometries. Int. J. Civil Struct. Eng. 2010, 1 (3), 591– 605,  DOI: 10.1056/NEJMoa2023184
      76
      Modeling of overall volumetric oxygen transfer by plunging jets of different geometries
      Singh, Shakti; Deswal, Surinder; Pal, Mahesh
      International Journal of Civil and Structural Engineering (2010), 1 (3), 591-605CODEN: IJCSKN; ISSN:0976-4399. (Integrated Publishing Association)
      The paper explores the potential of computational techniques in the modeling of overall volumetric oxygen transfer by plunging jets of different geometries, namely circular, square, rectangular and rectangular with rounded edge. The results predicted from neural network, support vector machines and Gaussian process techniques are compared in terms of correlation coeff., root mean square error and coeff. of detn. The outcome suggests the utility of all these computational techniques in the designing and performance evaluation of plunging jets oxygenation systems of different geometric shapes. However, support vector machines have performed better in comparison to other techniques. It has predicted overall volumetric oxygen transfer coeff. with a correlation coeff. of 0.985, coeff. of detn. of 0.968 and root mean square error of 0.002. Further, the scattering (within ±15%) is lowest in case of support vector machines approach. A comparison of results suggests that support vector machines works well and can be successfully used in modeling oxygen transfer by plunging jets of different geometries and configurations.
    77. 77
      Sihag, P.; Singh, B.; Sepah Vand, A.; Mehdipour, V. Modeling the infiltration process with soft computing techniques. ISH J. Hydraul. Eng. 2020, 26 (2), 138– 152,  DOI: 10.1080/09715010.2018.1464408
      There is no corresponding record for this reference.
    78. 78
      Mahdiyar, A.; Armaghani, D. J.; Marto, A.; Nilashi, M.; Ismail, S. Rock tensile strength prediction using empirical and soft computing approaches. Bull. Eng. Geol. Environ. 2019, 78, 4519– 4531,  DOI: 10.1007/s10064-018-1405-4
      There is no corresponding record for this reference.
    79. 79
      Malik, A.; Kumar, A. Pan evaporation simulation based on daily meteorological data using soft computing techniques and multiple linear regression. Water Resour. Manage. 2015, 29, 1859– 1872,  DOI: 10.1007/s11269-015-0915-0
      There is no corresponding record for this reference.
    80. 80
      Armaghani, D. J.; Asteris, P. G. A comparative study of ANN and ANFIS models for the prediction of cement-based mortar materials compressive strength. Neural Comput. Appl. 2021, 33 (9), 4501– 4532,  DOI: 10.1007/s00521-020-05244-4
      There is no corresponding record for this reference.