Abstract

This study aims to evaluate the learning ability and performance of five meta-heuristic optimization algorithms in training forward and recurrent fuzzy-based machine learning models, such as adaptive neuro-fuzzy inference system (ANFIS) and RANFIS (recurrent ANFIS), to predict hydraulic jump characteristics, i.e., downstream flow depth (h2) and jump length (Lj). To meet this end, the firefly algorithm (FA), particle swarm algorithm (PSO), whale optimization algorithm (WOA), genetic algorithm (GA), and moth-flame optimization algorithm (MFO) are embedded with the fuzzy-based models, which represent the main contribution of this study. The analysis of the results of predicting hydraulic jump characteristics shows that the embedded ANFIS and RANFIS models are more accurate than the empirical relations proposed by the previous researchers. Comparing the performance of the embedded RANFISs and ANFISs methods in predicting Lj represents the superiority of the RANFIS models to the ANFISs. The results of the sensitivity analysis show that among the input independent parameters, flow discharge (Q) is the most important factor in predicting downstream flow depth in weak, oscillating, and steady hydraulic jumps (1.7 < Froude number < 9), while the upstream flow depth (h1) is more important than the other input parameters in strong hydraulic jumps (Froude number > 9).

HIGHLIGHTS

  • Modeling hydraulic jump using integrative soft computing techniques.

  • Appraisal of forward and recurrent fuzzy-based models in predicting hydraulic jump characteristics.

  • Application of five heuristic algorithms, including FA, PSO, WOA, GA, and MFO.

  • Sensitivity analyses for the weak, oscillating, steady, and strong hydraulic jumps.

Notation

     
  • = model output

  •  
  • = observational value of a phenomenon

  •  
  • B

    = flume width

  •  
  • C1

    = cognitive acceleration (related to the PSO method)

  •  
  • C2

    = social acceleration (related to the PSO method)

  •  
  • CRMS

    = centered root-mean-square difference

  •  
  • d50

    = median diameter of the base material

  •  
  • Fr

    = the Froude number

  •  
  • g

    = the gravitational acceleration

  •  
  • h1

    = upstream supercritical flow depth

  •  
  • h2

    = downstream flow depth

  •  
  • h2*

    = downstream flow depth calculated from the RANFIS model

  •  
  • hc

    = critical depth

  •  
  • I1, I2, … ,Ik

    = independent input parameters in RANFIS

  •  
  • IA

    = index of agreement

  •  
  • Ks

    = roughness heights

  •  
  • Lj

    = hydraulic jump length

  •  
  • Lr

    = length roller in the hydraulic jump

  •  
  • MAE

    = mean absolute error

  •  
  • N

    = number of data pieces

  •  
  • NSE

    = the Nash–Sutcliffe efficiency

  •  
  • Q

    = flow discharge

  •  
  • r

    = correlation coefficient

  •  
  • R

    = hydraulic radius

  •  
  • R2

    = coefficient of determination

  •  
  • RMSE

    = root mean square error

  •  
  • S0

    = slope of the bed

  •  
  • SD

    = standard deviation

  •  
  • u*

    = shear velocity

  •  
  • υ

    = kinematic viscosity

INTRODUCTION

Description

In open channels, depends on the ratio of inertial and gravitational forces, free-surface flow is divided into three groups, including subcritical, critical, and supercritical. Subsequently, the flow in open channels (such as rivers and spillways) can be converted from subcritical to supercritical, or from supercritical to subcritical. The depth of the flow from subcritical to supercritical often changes smoothly by passing the critical depth. A sudden transition from supercritical flow to subcritical flow is associated with a high level of turbulence and energy loss and occurs in a relatively short interval; a phenomenon called the hydraulic jump.

A hydraulic jump is a complex hydraulic phenomenon in which knowing its characteristics such as the length (Lj) and secondary depth (h2) is a necessity for appropriate planning of different types of hydraulic structures such as stilling basins (Hager 1992; Montano & Felder 2020). In the current study, soft computing techniques are applied to predict Lj and h2 parameters. In the following subsections, the relevant studies are mentioned and, thereafter, the objectives and contribution of the research are pointed out clearly.

Literature review

Many studies have addressed modeling hydraulic jump phenomenon using different approaches such as laboratory studies, hard computing simulations (e.g., numerical simulation), and soft computing techniques (Pagliara et al. 2008; Gerami Moghadam et al. 2019; Gu et al. 2019). In terms of experimental and numerical perspective, hydraulic jumps have been broadly investigated and studied over the last decades. In Table 1, a summary of some pertinent experimental and numerical studies is provided.

Table 1

A summary of reported studies on modeling hydraulic jump characteristics using experimental and numerical studies

ApproachAuthor(s)/yearParameterObjective of the studyConclusion remarks
Experimental Silvester (1964)  Lj Estimating the hydraulic jump length (Lj) of horizontal channels A semi-empirical solution was provided for the jump length 
Experimental Hughes & Flack (1984)  Lj, h2 Investigating properties of the hydraulic jump over rough beds It was concluded that bed roughness diminishes both the length and the depth of a hydraulic jump 
Experimental Hager et al. (1990)  Lr Defining the length roller in the classical hydraulic jump (LrRelations for design were proposed based on experiments conducted in three different channels 
Experimental Mohamed Ali (1991)  Lj Analyzing the influence of stilling basins on the length of the jump Providing a general formula for the length of jump on a rough bed 
Numerical Ma et al. (2001)  Lj, h2, Lr Using kε turbulence model for numerical investigation of the characteristics of submerged hydraulic jumps Providing information regarding the turbulent structure of the hydraulic jump 
Experimental Pagliara et al. (2008)  h2 Investigation of the parameters that influence the sequent flow depths and modeling the length of the hydraulic jump The experimental data were analyzed to extract a formulation of the correction coefficient based in the bed roughness 
Experimental Abbaspour et al. (2009)  Lj, h2 Studying the hydraulic jump properties effected on corrugated beds The results indicated that the length of the jump and the downstream flow depth on corrugated beds are smaller compared to the jumps on smooth bed 
Experimental Pagliara & Palermo (2015)  Lj, h2 Studying the hydraulic jump characteristics in rough adverse-sloped channels A semi-theoretical predictive relationship was proposed to estimate jump characteristics for a wide range of hydraulic and geometric conditions covering both rough and smooth beds 
Numerical Bayon et al. (2016)  Lr, h2 Challenging the capability of two numerical models for simulating the hydraulic jump Both numerical models gave promising results compared to the experimental observations 
Experimental Pourabdollah et al. (2019)  h2 Experimental investigation of hydraulic jump characteristics on various beds, slopes, and step heights In addition to presenting the observed values, two analytical solutions were also developed based on the momentum equation 
Numerical Gu et al. (2019)  Lj, h2 Using the SPH (smoothed particle hydrodynamics) meshless method to simulate the hydraulic jump on corrugated beds Numerical simulations were accurate in modeling different hydraulic aspects of the hydraulic jumps 
ApproachAuthor(s)/yearParameterObjective of the studyConclusion remarks
Experimental Silvester (1964)  Lj Estimating the hydraulic jump length (Lj) of horizontal channels A semi-empirical solution was provided for the jump length 
Experimental Hughes & Flack (1984)  Lj, h2 Investigating properties of the hydraulic jump over rough beds It was concluded that bed roughness diminishes both the length and the depth of a hydraulic jump 
Experimental Hager et al. (1990)  Lr Defining the length roller in the classical hydraulic jump (LrRelations for design were proposed based on experiments conducted in three different channels 
Experimental Mohamed Ali (1991)  Lj Analyzing the influence of stilling basins on the length of the jump Providing a general formula for the length of jump on a rough bed 
Numerical Ma et al. (2001)  Lj, h2, Lr Using kε turbulence model for numerical investigation of the characteristics of submerged hydraulic jumps Providing information regarding the turbulent structure of the hydraulic jump 
Experimental Pagliara et al. (2008)  h2 Investigation of the parameters that influence the sequent flow depths and modeling the length of the hydraulic jump The experimental data were analyzed to extract a formulation of the correction coefficient based in the bed roughness 
Experimental Abbaspour et al. (2009)  Lj, h2 Studying the hydraulic jump properties effected on corrugated beds The results indicated that the length of the jump and the downstream flow depth on corrugated beds are smaller compared to the jumps on smooth bed 
Experimental Pagliara & Palermo (2015)  Lj, h2 Studying the hydraulic jump characteristics in rough adverse-sloped channels A semi-theoretical predictive relationship was proposed to estimate jump characteristics for a wide range of hydraulic and geometric conditions covering both rough and smooth beds 
Numerical Bayon et al. (2016)  Lr, h2 Challenging the capability of two numerical models for simulating the hydraulic jump Both numerical models gave promising results compared to the experimental observations 
Experimental Pourabdollah et al. (2019)  h2 Experimental investigation of hydraulic jump characteristics on various beds, slopes, and step heights In addition to presenting the observed values, two analytical solutions were also developed based on the momentum equation 
Numerical Gu et al. (2019)  Lj, h2 Using the SPH (smoothed particle hydrodynamics) meshless method to simulate the hydraulic jump on corrugated beds Numerical simulations were accurate in modeling different hydraulic aspects of the hydraulic jumps 

In accordance with the topic of this paper, pertinent published studies regarding the estimation/prediction of hydraulic jump characteristics (e.g., the jump length) using soft computing methods are mentioned. Abbaspour et al. (2013) employed artificial neural networks (ANNs) and genetic programming (GP) for the estimation of hydraulic jump characteristics such as the free-surface location and energy dissipation. The findings of the study showed that the estimations of the ANN and GP models were in good agreement with the measured data. Additionally, the results of the ANN and GP models were compared. It was found that the ANNs gave better results than GP models.

Karbasi & Azamathulla (2016) attempted to model the properties of hydraulic jumps over rough beds with the aid of gene expression programming (GEP) models. The performances of the GEP model were compared to empirical equations, ANN, and support vector regression (SVR). A comparison of the obtained results showed that the ANN and SVR models outperformed the GEP. Azimi et al. (2018a) predicted the length of hydraulic jump on rough beds utilizing an integrated model of adaptive neuro-fuzzy inference system (ANFIS) and firefly algorithm (ANFIS-FA). Kumar et al. (2019) applied the ANN to predict the sequent depth ratio of hydraulic jump on the sloping floor with rounded and crushed aggregates. The results indicated that the ANN model has the capability to predict the sequent depth ratio with the appropriate accuracy. For the sake of the building ANIFS structure, several effective parameters, including the relative roughness of the jump, Froude number (Fr), and sequent depth were analyzed. Outcomes of the study revealed that the applied integrated ANFIS-FA models were superior to the standard ANFIS model for estimating the length of the hydraulic jump. In Table 2, a summary of the published reports regarding soft computing methods in simulating hydraulic jump is given.

Table 2

A summary of reported studies on modeling hydraulic jump characteristics using soft computing techniques

Soft computing method(s)Author(s)/yearParameter(s)Objective of the studyConclusion remarks
MLPNN Omid et al. (2005)  Lj, h2 An artificial neural network (ANN) approach was applied to model sequent depth and jump length For the rectangular section, the neural network model successfully predicted the jump length as well as the sequent depth values 
MLPNN Naseri & Othman (2012)  Lj In this study, an ANN technique was developed to determine the length of the hydraulic jumps in a rectangular section with a horizontal apron A comparison between the selected ANN model and the empirical Silvester equation was also made, and the results showed that the ANN method was more precise 
MLPNN/GP Abbaspour et al. (2013)  Lj, h2 ANNs and genetic programming (GP) were used for the estimation of hydraulic jump characteristics Results showed that the proposed ANN models were much more accurate than the GP models 
MLPNN/GRNN Houichi et al. (2013)  Lj Two different ANNs were implemented to model the relative lengths of hydraulic jumps The results demonstrated that both the MLPNN and GRNN were reliable predictive tools for simulating the hydraulic jump properties 
GEP/SVR/MLPNN Karbasi & Azamathulla (2016)  Lj Application of several soft computing models to predict characteristics of hydraulic jumps over rough beds ANN and SVR provided better results than the GEP model 
ANFIS/ANFIS-FA Azimi et al. (2018a)  Lr Evaluating the potential of FA algorithm in simulating the hydraulic jump Integrating the FA algorithm with ANFIS made the standard ANFIS produce more accurate results 
GMDH/MLPNN Azimi et al. (2018b)  Lr Estimating the roller length of hydraulic jumps on rough beds using GMDH and ANN models The suggested soft computing models’ predictions were closer to the observed values than a number of other empirical models 
ANFIS/Differential Evolution Gerami Moghadam et al. (2019)  Lj A hybrid method (ANFIS-DE) was proposed for modeling hydraulic jumps on sloping rough beds Two parameters including the ratio of sequent depths and the Froude number were identified as the most important parameters in modeling the hydraulic jump length 
GEP Azimi et al. (2019)  Lr Prediction of the roller length of a hydraulic jump A simple and practical equation was proposed for predicting the length of a hydraulic jump 
MLPNN Kumar et al. (2019)  h2/h1 Prediction of sequent depth ratio MLPNN provided better results than empirical models 
ELM Azimi et al. (2020)  Lj Prediction of hydraulic jump length on slope rough beds The flow Froude number at upstream was introduced as the most effective parameter in predicting the jump length 
Soft computing method(s)Author(s)/yearParameter(s)Objective of the studyConclusion remarks
MLPNN Omid et al. (2005)  Lj, h2 An artificial neural network (ANN) approach was applied to model sequent depth and jump length For the rectangular section, the neural network model successfully predicted the jump length as well as the sequent depth values 
MLPNN Naseri & Othman (2012)  Lj In this study, an ANN technique was developed to determine the length of the hydraulic jumps in a rectangular section with a horizontal apron A comparison between the selected ANN model and the empirical Silvester equation was also made, and the results showed that the ANN method was more precise 
MLPNN/GP Abbaspour et al. (2013)  Lj, h2 ANNs and genetic programming (GP) were used for the estimation of hydraulic jump characteristics Results showed that the proposed ANN models were much more accurate than the GP models 
MLPNN/GRNN Houichi et al. (2013)  Lj Two different ANNs were implemented to model the relative lengths of hydraulic jumps The results demonstrated that both the MLPNN and GRNN were reliable predictive tools for simulating the hydraulic jump properties 
GEP/SVR/MLPNN Karbasi & Azamathulla (2016)  Lj Application of several soft computing models to predict characteristics of hydraulic jumps over rough beds ANN and SVR provided better results than the GEP model 
ANFIS/ANFIS-FA Azimi et al. (2018a)  Lr Evaluating the potential of FA algorithm in simulating the hydraulic jump Integrating the FA algorithm with ANFIS made the standard ANFIS produce more accurate results 
GMDH/MLPNN Azimi et al. (2018b)  Lr Estimating the roller length of hydraulic jumps on rough beds using GMDH and ANN models The suggested soft computing models’ predictions were closer to the observed values than a number of other empirical models 
ANFIS/Differential Evolution Gerami Moghadam et al. (2019)  Lj A hybrid method (ANFIS-DE) was proposed for modeling hydraulic jumps on sloping rough beds Two parameters including the ratio of sequent depths and the Froude number were identified as the most important parameters in modeling the hydraulic jump length 
GEP Azimi et al. (2019)  Lr Prediction of the roller length of a hydraulic jump A simple and practical equation was proposed for predicting the length of a hydraulic jump 
MLPNN Kumar et al. (2019)  h2/h1 Prediction of sequent depth ratio MLPNN provided better results than empirical models 
ELM Azimi et al. (2020)  Lj Prediction of hydraulic jump length on slope rough beds The flow Froude number at upstream was introduced as the most effective parameter in predicting the jump length 

MLPNN: multi-layer perceptron neural network; GP: genetic programming; GEP: Gene expression programming; SVR: support vector regression; GMDH: group method of data handling; ANFIS: adaptive neuro-fuzzy inference system; ELM: Extreme Learning Machine.

Objectives, novelty, and contribution

Analyzing the related subjects associated with simulating hydraulic jump using soft computing methods implies that the neural network-based models (ANNs and ANFIS) acted better than other types of soft computing methods (e.g., GEP, SVR, and GP). Besides, in recent years, the adaptive neural fuzzy inference system (ANFIS) method has been successfully used in modeling hydraulic phenomena such as scour depth at piers (Muzzammil & Ayyub 2010), daily streamflow (Li et al. 2018), spillways aerator air demand (Mahdavi-Meymand et al. 2019), and scour depth around pipelines (Sharafati et al. 2020a). Therefore, this method was selected for the purpose of the present study. Indeed, Azimi et al. (2018b) stated that integrating ANFIS with meta-heuristic algorithms would increase the competence of the standard ANFIS model. As a result, the main focus of this study is laid on using different types of integrated ANFIS models in predicting hydraulic jump characteristics over rough beds.

As well, several researchers have proved that the use of recurrent machine learning models would improve the performance of the counterpart standard ones (Bhattacharjee & Tollner 2016; Murali et al. 2020; Wei et al. 2020). In other words, considering a calculated output parameter as an input parameter to the soft computing model (here h2 as the secondary depth of hydraulic jump) can yield more precise results of the target parameter (here Lj as the length of the hydraulic jump).

Taking into account the above-mentioned issues, the contribution of the present research can be presented into three categories. Firstly, to the best knowledge of the authors, the recurrent soft computing models have not been already applied to model the hydraulic jump phenomenon. To this end, a recurrent intelligent method was employed, and results were compared with those of the standard model. In this paper, RANFIS (recurrent ANFIS) will be applied to model the hydraulic jump length (Lj) for the first time.

Second, the use of five different types of meta-heuristic optimization algorithms in learning ANFIS and RANFIS methods and comparing their performance will be accomplished. Genetic algorithm (GA), firefly algorithm (FA), and particle swarm optimization (PSO) algorithms are well-known methods with their successful applications reported by different researchers in engineering studies, especially hydraulics (Zounemat-Kermani et al. 2019a; Sharafati et al. 2020b; Zanganeh 2020). In this study, the performance of moth-flame optimization (MFO), and whale optimization algorithm (WOA), as two of the more recent intelligent algorithms, is additionally compared to those of other algorithms. So far, the capability of these five algorithms has not been assessed in engineering problems.

The third contribution of this study is associated with performing a detached sensitivity analysis in addition to the standard sensitivity analysis of the effective parameters, including discharge (Q) upstream supercritical flow depth (h1), roughness heights (Ks), and flume width (B) by detaching the hydraulic jump into four types (weak jump, oscillating jump, steady jump, and strong jump) based on the Froude number. As an illustration, the influence of discharge is on the length of the oscillating wave is far more important than other types of jumps.

The rest of the paper is organized as follows. In the next section (Methodology section), general properties of the hydraulic jump are briefly introduced. Then, the applied empirical equations, as well as the developed soft computing methods, are described. Section ‘Model implication and result’ provides the mathematical and soft computing predicted results for the downstream flow depth (h2) and hydraulic jump length (Lj) using the statistical measures and sensitivity analysis. Subsequently, in the ‘Discussion’ section, the Kruskal–Wallis test is used to check the significant difference between the predicted results. Finally, the ‘Conclusions’ section summarizes the general findings of the study applied.

METHODOLOGY

Hydraulic jump

The main characteristics of a hydraulic jump can be named as the upstream supercritical flow depth (h1), the downstream subcritical flow depth (h2), and the length of the hydraulic jump (the distance between the two parts (Lj)). These parameters are schematically shown in Figure 1.

Figure 1

Schematic section plan of hydraulic jump: Lr, Roller Length; Lj, the jump length; h1, upstream flow depth; h2, downstream flow depth.

Figure 1

Schematic section plan of hydraulic jump: Lr, Roller Length; Lj, the jump length; h1, upstream flow depth; h2, downstream flow depth.

In Figure 1, Ks is the roughness height of the particles in the channel bed and Lr is the roller length. As noted, in order to have a proper design of hydraulic structures like stilling basins downstream of the hydraulic jump, having the geometric characteristics of the jump is essential. In this regard, preliminary studies were carried out to provide a relationship between hydraulic jump parameters in horizontal channels with a smooth bed (Bakhmeteff 1932).

In this regard, and based on the general form of the momentum equation, the general equation for the free-surface hydraulic jump (also known as the Bélanger equation) was obtained. This equation is as follows:
formula
(1)
where Fr1 is the supercritical Froude number of flow at the upstream and are the ratio of sequent depths.

The phenomenon of hydraulic jump on smooth beds has been widely studied and reported by scholars such as Peterka (1958), Rajaratnam (1967), McCorquodale (1986), and Hager (1992).

One approach to determine either the smoothness or roughness of the channel bed is taking into account the thickness of the boundary layer or Reynold's number of the particle. The thickness of the boundary layer (δ) is calculated using the following formula:
formula
(2)
where υ is the kinematic viscosity, u* is shear velocity, R is the hydraulic radius, g is the gravitational acceleration, and S0 is the slope of the bed. If the value of bed roughness is greater than six times the thickness of the boundary layer, the bed is considered to be rough (Gadbois & Wilkerson 2014). Here, Reynold's number greater than 70 suggests the presence of a rough flow (Singh et al. 2007). However, studies of the hydraulic jump on rough beds have been addressed less than those on the smooth beds.

The preliminary researches on hydraulic jump on the rough beds were carried out by Rajaratnam (1968). The overall results of the researches indicated that the hydraulic jump on the rough bed creates a smaller downstream subcritical flow depth than that in which the bed is smooth (Carollo et al. 2007). Experimental studies indicated that the downstream depth of the jump on the rough bed is virtually half of the smooth bed (Ead & Rajaratnam 2002).

Empirical equations

Several researchers conducted studies for representing empirical relations for calculating sequent depths of hydraulic jump. Carollo et al. (2007) studied the hydraulic jump on several rough horizontal beds and presented a novel relation for the momentum equation for the ratio of sequent depths and the roller lengths. Pagliara et al. (2008) tested and analyzed the hydraulic jump properties on a rough bed and provided experimental relationships for these characteristics. In this section, the five empirical relations for predicting the downstream subcritical flow depth (h2) and one empirical relation for estimating the length of the jump are presented in Table 3.

Table 3

The empirical equations for estimating the h2 (downstream flow depth) and Lj (the jump length) used in this study

EquationPresented byRemarks
 Leutheusser & Kartha (1972)  The general quadratic equation for hydraulic jump 


α = 7.43, based on Carollo & Ferro (2004)  
Govinda Rao & Ramaprasad (1996)  A developed form of the general jump equation 
 Carollo & Ferro (2004)  Jump on rough horizontal beds 
 Carollo et al. (2007)  Jump on horizontal rough beds 
 Pagliara & Palermo (2015)  Flows over rough beds 
 Silvester (1964)  Hydraulic jump over horizontal channels 
EquationPresented byRemarks
 Leutheusser & Kartha (1972)  The general quadratic equation for hydraulic jump 


α = 7.43, based on Carollo & Ferro (2004)  
Govinda Rao & Ramaprasad (1996)  A developed form of the general jump equation 
 Carollo & Ferro (2004)  Jump on rough horizontal beds 
 Carollo et al. (2007)  Jump on horizontal rough beds 
 Pagliara & Palermo (2015)  Flows over rough beds 
 Silvester (1964)  Hydraulic jump over horizontal channels 

In the aforementioned equations, is the upstream Froude number, h1 and h2 are upstream supercritical and downstream subcritical flow depth, denotes the roughness height, d50 is the median diameter of the base material, hc is the critical depth, and S0 is the channel bed slope.

Soft computing techniques

In this study, the methods employed to predict the secondary depth are divided into two groups: soft computing based (ANFIS models) and empirical ones (see Table 3). Moreover, the methods for predicting the length of the jump are divided into two groups: soft computing based (ANFIS and RANFIS models) and one empirical relation (see Table 3). Today, many optimization algorithms have been developed to determine the relative and absolute extrema of complex functions. In the present work, five optimization algorithms were employed for training ANFIS and RANFIS, i.e., three frequently used algorithms (GA, PSO, and FA) and two novel ones (MFO and WOA). In the following, these methods are briefly explained.

ANFIS and recurrent ANFIS models

The ANFIS is an artificial intelligence learning algorithm that was first introduced by Jang (1993). ANFIS is a soft computing model generated by the learning ability of ANNs and the fuzzy inference system (FIS). By applying the fuzzy rules and sets of membership function (MFs), ANFIS is trained in accordance with the input–output data pairs of the available problem. The structure of ANFIS consists of five consequent layers, including the (1) fuzzification, (2) applying the fuzzy rules, (3) normalizing the MFs, (4) executing the consequent part of the fuzzy rules, and (5) defuzzification. Figure 2 shows the general procedure for setting up the developed ANFIS models.

Figure 2

Schematic plan for the architecture of the recurrent ANFIS (RANFIS) model applied.

Figure 2

Schematic plan for the architecture of the recurrent ANFIS (RANFIS) model applied.

Although the high capability and robustness of feed-forward ANFIS models have been proven by numerous researchers (Rajaee et al. 2009; Zounemat-Kermani & Scholz 2013); however, as a result of their feed-forward structure, a major deficiency of ANFIS models is that their effectiveness is limited to static problems. Thus, ANFISs are less efficient for representing dynamic processes in comparison with recurrent networks such as NARX models (Zounemat-Kermani et al. 2019b). Hence, it would be reasonable to upgrade the typical feed-forward ANFIS model to a recurrent version (RANFIS) for promoting its performance in modeling recurrent problems. In this paper, as well as employing the standard feed-forward ANFIS model for predicting the downstream subcritical flow depth of jump (h2) and the jump length (Lj), the jump length is also predicted with the predicted values for h2 using an RANFIS model. Figure 2 shows the constructed RANFIS structure with four independent inputs and one recurrent input in this study.

In this study, I1, I2, … , Ik denote the independent input parameters such as Q, Ks, B, and h1.

In this study, for converting the feed-forward ANFIS architecture to the recurrent ANFIS (RANFIS) type in predicting the jump length, the output of typical ANFIS will be changed to for the RANFIS. As shown in Figure 2, the output of the feed-forward ANFIS (h2) is fed-back to the input vector of the network.

Particle swarm optimization algorithm

The PSO proposed by Kennedy & Eberhart (1995) is a general random (stochastic) optimization method based on swarms such as the social behavior of birds, herds, and insects. In this method, a group of animals/insects randomly search for food in a space in which only one particle of food exists. None of the birds knows the location of the food. One of the best strategies to solve such problems is to follow birds that have the shortest distance from the food. This strategy is, in fact, the basis of the PSO algorithm. The main advantage of this method over other optimization strategies is the presence of a large number of swarming particles in it that leads to its flexibility against the problem of local optima entrapment. Instead of using mutation, this method exchanges information among the members of the population, i.e., particles in the group. In fact, each particle regulates its path with respect to its previous best location, and the previous best location obtained by its neighbors (Hang et al. 2016).

Firefly algorithm

FA was introduced by Yang (2010) as a multi-factor solution to difficult optimization problems. In this algorithm, the firefly attracts other fireflies using light signals. The artificial firefly defined in this algorithm is unisex, and one firefly can attract all other fireflies. The attraction is proportional to the level of brightness of each firefly; in other words, a firefly with less brightness is less attractive, and the one with more brightness is more attractive. Fireflies with less brightness are attracted by those with higher brightness. However, if there is no firefly brighter than the present firefly, the movement of fireflies will be random. This level of brightness is defined based on the objective function: Any firefly that maximizes/minimizes the objective function based on the problem requirements has further brightness and, thus, higher attraction. On the other hand, any firefly that minimizes/maximizes the problem has less brightness and, therefore, less attraction compared to others (Zhang et al. 2018).

Genetic algorithm

GA was first proposed by Holland (1975). This algorithm belongs to the group of random optimization algorithms that are proper for the optimization of complex problems with unknown search space. The main idea behind GA is Darwin's theory of evolution. This theory proposes that natural attributes that are more compatible with natural rules have a higher survival chance. Since GA does not have constraining assumptions regarding the search space, it is an effective and useful method for solving optimization problems (Cheong et al. 2017). In GA, first, a random set of solutions is generated as the initial population that is replaced by a new candidate in each generation. In each iteration of the algorithm, the population is evaluated by the objective function, and a number of the best candidates are selected for the following generation, creating a new population. In the next step, a number of members of this population are used for generating new children by genetic operators such as crossover and mutation. These steps are iterated until the algorithm reaches the appropriate solution.

Moth-flame optimization algorithm

MFO algorithm was introduced by Mirjalili (2015) inspired by moths. Moths show an interesting method for flying at night using moonlight. Their orientation mechanism is called the transverse orientation. For flight on a straight path, moths move by maintaining a fixed angle relative to the moon. Because the moon is far from them, the path will be straight. However, it is usually observed that moths move in a spiral form around artificial lights because of the weakness in transverse orientation when the flame is near. Thus, this method is useful only when the flame is located far away. In the MFO algorithm, primary solutions are the population of moths, and the same number of flames is also considered, forming two equal-sized matrices. The fitness of each matrix is determined using the introduced objective function. In this algorithm, moths and flames are both solutions to the problem that are different in terms of the method employed for updating them in each iteration. In fact, moths move in the search space, but flames are the best locations (i.e., solutions) of moths until the performed iteration. Also, the new position of moths is calculated using a spiral.

Whale optimization algorithm

WOA is inspired by humpback whales. The behavior of these whales while hunting, which is called the bubble-net feeding method, was mathematically modeled by Mirjalili & Lewis (2016), and the WOA algorithm was extracted. Humpback whales prefer to hunt groups of plankton and small fish near the water surface. Hunting is performed by creating bubbles along the diameter of a circle or other forms. WOA considers the best solution from the primary population based on the objective function as the location of the prey or its surroundings. The other whales update their position toward this position. The location of the solution must be updated at the end of each iteration of the algorithm. WOA simulates exploitation and an exploration phase. The exploitation phase consists of creating spiral bubbles around the prey and moving in this path toward the prey. The exploration phase comprises random searching for the prey (Mafarja & Mirjalili 2016). In this algorithm, instead of using a random method for discovering the prey, the location of the best solution is used to update new solutions.

Development of integrated ANFIS and recurrent ANFIS models

The standard version of ANFIS uses a combination of the gradient descent (GD) and least squares (LS) methods for adjusting the premise parameters (MFs parameters) and consequent parameters (linear parameters) for hybrid training the network in the forward and backward stages. The LS method in the forward stage is employed to adjust the consequent parameters (coefficients of the linear relation in Layer 4; see Figure 2), and then the back-propagation GD method in the backward stage is employed to determine the optimal premise parameter values (e.g., the slope, center, and width of the bell-shaped function of MFs in Layer 1; see Figure 2).

In this study, in addition to the application of a standard version of ANFIS/RANFIS models, the training parameters are also optimized by using five heuristic algorithms, including PSO, GA, FA, WOA, and MFO. From now on, these integrated models will be named as ANFIS-PSO/RANFIS-PSO, ANFIS-GA/RANFIS-GA, ANFIS-FA/RANFIS-FA, ANFIS-WOA/RANFIS-WOA, and ANFIS-MFO/RANFIS-MFO in this paper. The general procedure of applying integrated ANFIS/RANFIS models is shown in Figure 3.

Figure 3

General procedure for setting up the integrated ANFIS/RANFIS models.

Figure 3

General procedure for setting up the integrated ANFIS/RANFIS models.

MODEL OPTIMIZATION AND IMPLEMENTATION

Data preparation

In total, 574 data series were extracted from three reliable published scientific reports for modeling the hydraulic jump. Hughes & Flack (1984) carried out experiments on artificial rough beds and measured the characteristics of hydraulic jumps. These experiments were conducted on horizontal and rectangular flume having 0.305 m width. Five different artificially roughened beds were tested. In those experiments, the Froude number was varied from 2.34 to 10.5 and a total of 196 data under different conditions were measured. Carollo et al. (2007) conducted experiments on hydraulic jumps with a rough bed in a rectangular horizontal flume and measured the attributes of hydraulic jump. The flume was 14.4 m long, 0.6 m deep, and 0.6 m wide, and the Froude number was varied from 1.87 to 9.89. The number of data pieces used by these researchers was 367. Moreover, Ead & Rajaratnam (2002) performed experiments on hydraulic jump on a corrugated bed. The flume width, depth, and length were 0.446, 0.60, and 7.6 m, respectively. In their experiments, the hydraulic jump characteristics were measured under two roughened beds. The Froude number of the experiments was varied from 4 to 10, and the number of data pieces was 11. Table 4 presents a summary of the data used.

Table 4

The data sources and variation ranges of data used in this study

ParameterReferences
Hughes & Flack (1984) Ead & Rajaratnam (2002) Carollo et al. (2007) 
Q (m3/s) [0.00934, 0.01472] [0.02274, 0.09232] [0.01736, 0.07316] 
h1 (m) [0.0125, 0.0384] [0.0254, 0.0508] [0.0111, 0.0709] 
h2 (m) [0.0780, 0.1487] [0.104, 0.31] [0.0898, 0.2345] 
Lj (m) [0.3962, 0.8839] [0.41, 1.29] [0.18, 0.9] 
B (m) 0.305 0.446 0.6 
Ks (mm) [0, 11.3] 13 and 22 [0, 32] 
Fr [2.34, 10.5] [3.995, 9.996] [1.87, 9.89] 
Number of data 196 11 367 
ParameterReferences
Hughes & Flack (1984) Ead & Rajaratnam (2002) Carollo et al. (2007) 
Q (m3/s) [0.00934, 0.01472] [0.02274, 0.09232] [0.01736, 0.07316] 
h1 (m) [0.0125, 0.0384] [0.0254, 0.0508] [0.0111, 0.0709] 
h2 (m) [0.0780, 0.1487] [0.104, 0.31] [0.0898, 0.2345] 
Lj (m) [0.3962, 0.8839] [0.41, 1.29] [0.18, 0.9] 
B (m) 0.305 0.446 0.6 
Ks (mm) [0, 11.3] 13 and 22 [0, 32] 
Fr [2.34, 10.5] [3.995, 9.996] [1.87, 9.89] 
Number of data 196 11 367 

Input data

In this study, to predict the length of the hydraulic jump on a rough bed, flow rate (Q), upstream supercritical flow depth (h1), roughness heights (Ks), and flume width (B) are calculated using the mentioned methods. In RANFIS, the downstream subcritical flow depth (h2) enters the model as an input after the preliminary prediction. These parameters are given in Figure 1. Table 5 presents the structure of the input vector of the models.

Table 5

The input–output structure of the soft computing models as well as empirical relations created for predicting the attributes of hydraulic jump

ModelInput parametersTarget
ANFIS Q, h1, B, Ks h2 
ANFIS Q, h1, B, Ks Lj 
RANFIS Q, h1, B, Ks, h2* Lj 
Silvester (1964)  Q, h1, B Lj 
Leutheusser & Kartha (1972)  Q, h1, B h2 
Govinda Rao & Ramaprasad (1996)  Q, h1, B, α h2 
Carollo & Ferro (2004)  Q, h1, B, Ks h2 
Pagliara & Palermo (2015)  Q, h1, B, Ks h2 
ModelInput parametersTarget
ANFIS Q, h1, B, Ks h2 
ANFIS Q, h1, B, Ks Lj 
RANFIS Q, h1, B, Ks, h2* Lj 
Silvester (1964)  Q, h1, B Lj 
Leutheusser & Kartha (1972)  Q, h1, B h2 
Govinda Rao & Ramaprasad (1996)  Q, h1, B, α h2 
Carollo & Ferro (2004)  Q, h1, B, Ks h2 
Pagliara & Palermo (2015)  Q, h1, B, Ks h2 

h*2 = downstream flow depth calculated from the RANFIS model.

Models setup

To create a hydraulic jump model, first, the collected data were randomized and then divided into three categories of training (70%), validation (15%), and test (15%) using the cross-validation method. Next, the network was trained using training data, and validation data were used to resolve the over-training problem. Finally, the model was evaluated by comparing the results of the model and observational data in the test category. In this study, in addition to standard models of ANFIS and RANFIS, five optimization algorithms were incorporated to train ANFIS and RANFIS methods. The general characteristics and tuning coefficients of these methods are given in Table 6.

Table 6

The applied tunning parameters for the five learning algorithms embedded with the ANFIS and RANFIS used in this study

Meta-Heuristic optimization methodParameterValueMaximum epoch
FA Mutation Coefficient 0.2 2,000 
Attraction Coefficient 
Light Absorption Coefficient 
Mutation Coefficient Damping Ratio 0.98 
Search Space Range [−10, 10] 
PSO Initial inertia weight 2,000 
Inertia Weight Damping Ratio 0.98 
Cognitive acceleration (C1
Social acceleration (C2
Search range [−10, 10] 
GA Crossover 0.6 2,000 
Mutation 0.8 
Selection Pressure 
Search Space Range [−10, 10] 
MFO Search Space Range [−10, 10] 2,000 
WOA Search Space Range [−10, 10] 2,000 
Meta-Heuristic optimization methodParameterValueMaximum epoch
FA Mutation Coefficient 0.2 2,000 
Attraction Coefficient 
Light Absorption Coefficient 
Mutation Coefficient Damping Ratio 0.98 
Search Space Range [−10, 10] 
PSO Initial inertia weight 2,000 
Inertia Weight Damping Ratio 0.98 
Cognitive acceleration (C1
Social acceleration (C2
Search range [−10, 10] 
GA Crossover 0.6 2,000 
Mutation 0.8 
Selection Pressure 
Search Space Range [−10, 10] 
MFO Search Space Range [−10, 10] 2,000 
WOA Search Space Range [−10, 10] 2,000 

MODELS' EVALUATION

Statistical measures

The statistical parameters used in this study for evaluating different models include root mean square error (RMSE), coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE), mean absolute error (MAE), and index of agreement (IA). If the model output is denoted with and the observational value of a phenomenon with , each parameter is calculated as follows:
formula
(3)
formula
(4)
formula
(5)
formula
(6)
formula
(7)
where N shows the number of data pieces, and Bar denotes the mean of variables. Statistical parameters used here can be divided into two general categories. The smaller (and closer to zero) the RMSE and MAE, the better the performance of the model; however, the closer IA, R2, and NSE to 1, the higher the precision of the model.

Diagnostic analysis

An effective method for identifying the efficiency of the used models is the application of the Taylor diagram, which is a method for depicting the errors of different models on one diagram. In this diagram, three statistical parameters, including centered root-mean-square difference (CRMS), correlation coefficient (r), and standard deviation (SD) of various models, can be observed (Taylor 2001). The closer the representative point of a method is to the point of observational data, the better its performance would be in simulating the phenomenon.

MODELS IMPLICATION AND RESULT

Results for downstream flow depth (h2) prediction

In this study, both empirical and soft computing models were applied to predict the secondary flow depth of hydraulic jump (h2). Results of training and validation steps for predicting the h2 using the standard ANFIS are presented in Table 7.

Table 7

Statistical indices using the applied soft computing models for modeling the secondary flow depth of hydraulic jump (h2) in the training and validation phases

StageMethodStatistical parameters
RMSE (m)R2MAE (m)NSEIA
Train ANFIS-FA 0.0038 0.988 0.0029 0.988 0.997 
ANFIS-GA 0.0042 0.984 0.0033 0.984 0.996 
ANFIS-PSO 0.005 0.978 0.0039 0.978 0.994 
ANFIS 0.0067 0.96 0.0051 0.96 0.99 
ANFIS-WOA 0.0076 0.95 0.0058 0.95 0.987 
ANFIS-MFO 0.0083 0.94 0.0061 0.94 0.984 
Validation ANFIS-FA 0.0042 0.981 0.0014 0.981 0.995 
ANFIS-GA 0.005 0.972 0.0017 0.972 0.993 
ANFIS-PSO 0.0053 0.97 0.0019 0.969 0.992 
ANFIS 0.0067 0.951 0.0024 0.951 0.987 
ANFIS-WOA 0.0076 0.936 0.0028 0.936 0.984 
ANFIS-MFO 0.0081 0.93 0.0028 0.928 0.982 
StageMethodStatistical parameters
RMSE (m)R2MAE (m)NSEIA
Train ANFIS-FA 0.0038 0.988 0.0029 0.988 0.997 
ANFIS-GA 0.0042 0.984 0.0033 0.984 0.996 
ANFIS-PSO 0.005 0.978 0.0039 0.978 0.994 
ANFIS 0.0067 0.96 0.0051 0.96 0.99 
ANFIS-WOA 0.0076 0.95 0.0058 0.95 0.987 
ANFIS-MFO 0.0083 0.94 0.0061 0.94 0.984 
Validation ANFIS-FA 0.0042 0.981 0.0014 0.981 0.995 
ANFIS-GA 0.005 0.972 0.0017 0.972 0.993 
ANFIS-PSO 0.0053 0.97 0.0019 0.969 0.992 
ANFIS 0.0067 0.951 0.0024 0.951 0.987 
ANFIS-WOA 0.0076 0.936 0.0028 0.936 0.984 
ANFIS-MFO 0.0081 0.93 0.0028 0.928 0.982 

Statistical measures calculated in Table 7 show that the modeling methods used here were properly trained using the laboratory input data. Even for the poorest method, R2 was higher than 0.9, RMSE and MAE were close to 0, and NSE and IA were close to 1. Results also demonstrate that FA, GA, and PSO increased the compatibility of ANFIS with laboratory results, whereas WOA and MFO lowered the efficiency of the integrated ANFIS models compared to the standard ANFIS. These results are noticeable for both training and validation data. Among these methods, ANFIS-FA estimated h2 in the training and validation phases with more precision. During training with this method, RMSE was the minimum (0.0038 m), and NSE was the maximum (0.988). Although the performance of methods matters in training and validation, the results of the testing phase are of special importance, serving as the main criterion for depicting the performance of various methods. Table 8 presents the results of different soft computing models (i.e., the standard ANFIS and its combination with different algorithms) and three empirical relations in predicting h2 in the test phase.

Table 8

Statistical indices using the applied soft computing and empirical models for predicting the secondary flow depth of hydraulic jump (h2) in the testing phase

MethodStatistical parameters
RMSE (m)R2MAE (m)NSEIARMSE improvement %RMSE improvement %
ANFIS-FA 0.0053 0.975 0.0016 0.971 0.993 0.871 30.263 
ANFIS-GA 0.006 0.967 0.0018 0.962 0.991 0.854 21.053 
ANFIS-PSO 0.0063 0.962 0.002 0.959 0.99 0.846 17.105 
ANFIS 0.0076 0.944 0.0026 0.94 0.985 0.815 Base 
ANFIS-WOA 0.0079 0.938 0.0027 0.934 0.984 0.807 −3.947 
ANFIS-MFO 0.0084 0.931 0.0028 0.926 0.982 0.795 −10.526 
Carollo & Ferro (2004)  0.0242 0.626 0.007 0.388 0.865 0.41 – 
Pagliara & Palermo (2015)  0.0339 0.632 0.0116 −0.203 0.785 0.173 – 
Govinda Rao & Ramaprasad (1996)  0.035 0.573 0.012 −0.316 0.768 0.146 – 
Leutheusser & Kartha (1972)  0.041 0.576 0.014 −0.794 0.722 Base – 
MethodStatistical parameters
RMSE (m)R2MAE (m)NSEIARMSE improvement %RMSE improvement %
ANFIS-FA 0.0053 0.975 0.0016 0.971 0.993 0.871 30.263 
ANFIS-GA 0.006 0.967 0.0018 0.962 0.991 0.854 21.053 
ANFIS-PSO 0.0063 0.962 0.002 0.959 0.99 0.846 17.105 
ANFIS 0.0076 0.944 0.0026 0.94 0.985 0.815 Base 
ANFIS-WOA 0.0079 0.938 0.0027 0.934 0.984 0.807 −3.947 
ANFIS-MFO 0.0084 0.931 0.0028 0.926 0.982 0.795 −10.526 
Carollo & Ferro (2004)  0.0242 0.626 0.007 0.388 0.865 0.41 – 
Pagliara & Palermo (2015)  0.0339 0.632 0.0116 −0.203 0.785 0.173 – 
Govinda Rao & Ramaprasad (1996)  0.035 0.573 0.012 −0.316 0.768 0.146 – 
Leutheusser & Kartha (1972)  0.041 0.576 0.014 −0.794 0.722 Base – 

Statistical analyses on the obtained results (Table 8) reveal that conventional ANFIS and its combination with optimization algorithms predict h2 with higher accuracy. Mean RMSE was 0.013 m for soft computing models, while it was 0.019 m for the best empirical relations. Based on the calculated statistical parameters, ANFIS-FA showed the highest performance in predicting the secondary depth of the jump. For this method, RMSE and MAE equaled 0.0053 and 0.0016 m, respectively (minimum values among the applied methods). Also, R2, NSE, and IA were 0.971, 0.975, and 0.993, respectively. PSO, FA, and GA increased the precision of ANFIS modeling by 30.263, 17.105, and 21.053%, respectively, whereas MFO and WOA reduced this precision by 10.526 and 3.947%, respectively. Another method for the analysis and evaluation of results is the use of scatter plots (Figure 4).

Figure 4

Scatter plots of the predicted versus measured downstream depths of the hydraulic jumps (h2) for the developed ANFIS models in the testing phase.

Figure 4

Scatter plots of the predicted versus measured downstream depths of the hydraulic jumps (h2) for the developed ANFIS models in the testing phase.

The scatter plots in Figure 4 depict the superior performance of soft computing models compared to experimental ones. Scatter points for these methods are closer to the 45° line compared to experimental methods. Scatter points for soft computing models are not located outside the −20% and +20% lines, while this is true for the points of some experimental methods. Scatter points for soft computing models are similar to each other in terms of dispersion, showing that they have similar performance. Still, the FA outperforms the others.

Another method to evaluate and visualize the results of various methods is plotting the Taylor diagram. Figure 5 illustrates this diagram for the predicting results of the applied models.

Figure 5

Comparing performances of the empirical and soft computing (ANFIS) models in predicting h2 using Taylor diagram.

Figure 5

Comparing performances of the empirical and soft computing (ANFIS) models in predicting h2 using Taylor diagram.

Based on Figure 5, the points belonging to ANFIS methods are closer to the points from the observation point, suggesting that the ANFIS and its combination with optimization algorithms outperform the empirical methods. Points belonging to soft computing models are close to each other, and some are overlapping, indicating that their output and performance are close. Still, the ANFIS-FA is closer to the observational point and outperforms the other methods.

Results for hydraulic jump length (Lj) prediction

Another parameter predicted in this research is the length of the hydraulic jump on the rough bed (Lj). This parameter is modeled using RANFIS, standard feed-forward ANFIS, and their combinations with the meta-heuristic algorithms. Table 9 presents the results of the training and validation phases for both the ANFIS and RANFIS models.

Table 9

Statistical indices using the applied soft computing models for modeling the hydraulic jump length (Lj) in the training and validation phases

StageANFIS
RANFIS
Statistical parameters
Statistical parameters
MethodRMSE (m)R2IARMSE (m)R2IA
Train ANFIS-PSO 0.0495 0.863 0.962 0.0497 0.861 0.962 
ANFIS-FA 0.0484 0.869 0.964 0.0506 0.857 0.961 
ANFIS-GA 0.0468 0.877 0.966 0.0501 0.859 0.961 
ANFIS 0.054 0.837 0.954 0.0562 0.823 0.949 
ANFIS-MFO 0.0577 0.813 0.946 0.064 0.771 0.931 
ANFIS-WOA 0.0705 0.721 0.914 0.0711 0.717 0.912 
Validation ANFIS-PSO 0.0432 0.887 0.97 0.0466 0.869 0.965 
ANFIS-FA 0.0498 0.849 0.959 0.0461 0.87 0.964 
ANFIS-GA 0.049 0.854 0.96 0.0531 0.829 0.95 
ANFIS 0.0527 0.831 0.953 0.0557 0.812 0.947 
ANFIS-MFO 0.0603 0.782 0.933 0.0661 0.737 0.916 
ANFIS-WOA 0.0748 0.661 0.89 0.0751 0.658 0.889 
StageANFIS
RANFIS
Statistical parameters
Statistical parameters
MethodRMSE (m)R2IARMSE (m)R2IA
Train ANFIS-PSO 0.0495 0.863 0.962 0.0497 0.861 0.962 
ANFIS-FA 0.0484 0.869 0.964 0.0506 0.857 0.961 
ANFIS-GA 0.0468 0.877 0.966 0.0501 0.859 0.961 
ANFIS 0.054 0.837 0.954 0.0562 0.823 0.949 
ANFIS-MFO 0.0577 0.813 0.946 0.064 0.771 0.931 
ANFIS-WOA 0.0705 0.721 0.914 0.0711 0.717 0.912 
Validation ANFIS-PSO 0.0432 0.887 0.97 0.0466 0.869 0.965 
ANFIS-FA 0.0498 0.849 0.959 0.0461 0.87 0.964 
ANFIS-GA 0.049 0.854 0.96 0.0531 0.829 0.95 
ANFIS 0.0527 0.831 0.953 0.0557 0.812 0.947 
ANFIS-MFO 0.0603 0.782 0.933 0.0661 0.737 0.916 
ANFIS-WOA 0.0748 0.661 0.89 0.0751 0.658 0.889 

Based on Table 9, in the training and validation phases, both ANFIS and RANFIS managed to model Lj accurately. The statistical parameters show that ANFIS and RANFIS were trained similarly, such that mean RMSE was 0.0569 and 0.0544 m, respectively, for RANFIS and ANFIS. The performance of the optimization algorithms was similar to their performance in the previous problem (i.e., predicting the secondary depth of hydraulic jump). GA, PSO, and FA improved training, while MFO and WOA degraded it. In this step of modeling, ANFIS and RANFIS models show a similar performance. Still, ANFIS was trained better than RANFIS. The results of the testing phase based on the comparison of methods are presented in Table 10.

Table 10

Statistical indices using the applied soft computing and empirical models for predicting the hydraulic jump length (Lj) in the testing phase

MethodStatistical parameters
RMSER2MAENSEIARMSE changes %RMSE improvement %
Recurrent ANFIS 
 RANFIS-GA 0.046 0.846 0.016 0.841 0.960 20.86 92.03 
 RANFIS-PSO 0.046 0.838 0.016 0.837 0.957 20 91.94 
 RANFIS-FA 0.047 0.835 0.015 0.834 0.955 19.14 91.86 
 RANFIS 0.053 0.789 0.018 0.787 0.940 8.45 90.78 
 RANFIS-MFO 0.065 0.685 0.022 0.678 0.906 −12.41 88.68 
 RANFIS-WOA 0.074 0.599 0.026 0.588 0.880 −27.24 87.19 
Conventional ANFIS 
 ANFIS-GA 0.052 0.799 0.017 0.796 0.944 10.34 90.97 
 ANFIS-PSO 0.047 0.834 0.016 0.831 0.956 18.96 91.84 
 ANFIS-FA 0.048 0.827 0.016 0.824 0.953 16.72 91.61 
 ANFIS 0.058 0.751 0.020 0.746 0.929 Base 89.93 
 ANFIS-MFO 0.064 0.692 0.022 0.687 0.909 −10.86 88.84 
 ANFIS-WOA 0.074 0.598 0.026 0.587 0.879 −27.24 87.19 
Empirical relation 
Silvester (1964)  0.576 0.296 0.229 −24.09 0.270 −893.10 Base 
MethodStatistical parameters
RMSER2MAENSEIARMSE changes %RMSE improvement %
Recurrent ANFIS 
 RANFIS-GA 0.046 0.846 0.016 0.841 0.960 20.86 92.03 
 RANFIS-PSO 0.046 0.838 0.016 0.837 0.957 20 91.94 
 RANFIS-FA 0.047 0.835 0.015 0.834 0.955 19.14 91.86 
 RANFIS 0.053 0.789 0.018 0.787 0.940 8.45 90.78 
 RANFIS-MFO 0.065 0.685 0.022 0.678 0.906 −12.41 88.68 
 RANFIS-WOA 0.074 0.599 0.026 0.588 0.880 −27.24 87.19 
Conventional ANFIS 
 ANFIS-GA 0.052 0.799 0.017 0.796 0.944 10.34 90.97 
 ANFIS-PSO 0.047 0.834 0.016 0.831 0.956 18.96 91.84 
 ANFIS-FA 0.048 0.827 0.016 0.824 0.953 16.72 91.61 
 ANFIS 0.058 0.751 0.020 0.746 0.929 Base 89.93 
 ANFIS-MFO 0.064 0.692 0.022 0.687 0.909 −10.86 88.84 
 ANFIS-WOA 0.074 0.598 0.026 0.587 0.879 −27.24 87.19 
Empirical relation 
Silvester (1964)  0.576 0.296 0.229 −24.09 0.270 −893.10 Base 

Based on Table 10, RANFIS has a higher precision in modeling Lj in comparison to ANFIS. Compared to ANFIS, RANFIS has a smaller RMSE and MAE (i.e., 0.0531 and 0.0178 m, respectively) and a higher NSE, R2, and IA (0.789, 0.787, and 0.94, respectively). Among the five meta-heuristic algorithms used for training, FA, PSO, and GA improved the performance of both ANFIS models, whereas MFO and WOA weakened it. Figure 6 presents the scatter plot for visualizing and comparing the performance of different methods.

Figure 6

Scatter plots of the predicted versus measured length of the hydraulic jumps (Lj) for the developed ANFIS and RANFIS models in the testing phase.

Figure 6

Scatter plots of the predicted versus measured length of the hydraulic jumps (Lj) for the developed ANFIS and RANFIS models in the testing phase.

Based on Figure 6, the applied methods show very close performances. The scattering of points shows that recurrent methods (i.e., RANFIS) have less dispersion and, therefore, better performance compared to standard ones (i.e., ANFIS). The Taylor diagram for the length of the hydraulic jump (Lj) predicting is depicted in Figure 7.

Figure 7

Comparing performances of the empirical, ANFIS and RANFIS models in predicting Lj using Taylor diagram.

Figure 7

Comparing performances of the empirical, ANFIS and RANFIS models in predicting Lj using Taylor diagram.

Based on Figure 7, although the points are close to each other, RANFIS-GA was closer to the observational point, suggesting its superior performance compared to others. GA, PSO, and FA improved the performance of RANFIS compared to ANFIS, while the performance of MFO and WOA had a slight effect on RANFIS and even worsened it.

Sensitivity analysis

In this step, the simple correlation coefficient (SCC) statistical parameter was employed to examine the effect of independent input parameters on modeling and perform sensitivity analysis. Based on the Froude number, the data used for this purpose are divided into four groups (Figure 8).

Figure 8

Various types of hydraulic jumps based on the Froude number.

Figure 8

Various types of hydraulic jumps based on the Froude number.

This classification was performed based on the type of jump. A Froude number between 1.7 and 2.5 is a weak jump, 2.5–4.5 is an oscillating jump, 4.5–9 is a steady jump, and >9 is a strong jump.

Based on Table 4, the size of particles and width of the channel of the collected data have slight diversity; therefore, the results in Table 11 on these two parameters are not reliable. In comparison, the results for Q and h1 are reliable. Based on Table 11, for the total data, h1 and Q have almost equal importance in Lj modeling, but Q is more significant than h1. For weak, oscillating, and steady jumps, the importance of Q is more than that of h1, while the opposite is true for strong jump (with very large Froude numbers). For the case of Lj, it is found that the importance of Q is reduced by increasing the Froude number while that of h1 is increased such that Q is more important than h1 for weak jump, and the opposite is true for steady and strong jumps.

Table 11

The results of sensitivity analysis for exploring the effectiveness of input parameters in different types of hydraulic jumps

ParameterHydraulic jump type
Nonsegregated
Weak jump
Oscillating jump
Steady jump
Strong jump
SCC-LjSCC-h2SCC-LjSCC-h2SCC-LjSCC-h2SCC-LjSCC-h2SCC-LjSCC-h2
Q (m3/s) 0.08 0.837 0.123 0.948 0.596 0.931 0.05 0.845 0.034 0.557 
h1 (m) −0.102 0.456 0.058 0.866 0.543 0.764 0.264 0.724 0.816 0.81 
B (m) −0.23 0.592 −0.126 0.676 0.232 0.742 −0.327 0.557 −0.526 0.006 
Ks (mm) 0.001 0.19 0.895 0.18 0.325 0.404 −0.288 0.125 −0.626 −0.122 
ParameterHydraulic jump type
Nonsegregated
Weak jump
Oscillating jump
Steady jump
Strong jump
SCC-LjSCC-h2SCC-LjSCC-h2SCC-LjSCC-h2SCC-LjSCC-h2SCC-LjSCC-h2
Q (m3/s) 0.08 0.837 0.123 0.948 0.596 0.931 0.05 0.845 0.034 0.557 
h1 (m) −0.102 0.456 0.058 0.866 0.543 0.764 0.264 0.724 0.816 0.81 
B (m) −0.23 0.592 −0.126 0.676 0.232 0.742 −0.327 0.557 −0.526 0.006 
Ks (mm) 0.001 0.19 0.895 0.18 0.325 0.404 −0.288 0.125 −0.626 −0.122 

DISCUSSION

In the previous sections, we discussed and analyzed the superior methods and parameters affecting the hydraulic jump phenomenon. Still, the question remains whether the employed methods present statistically different results for various models. To answer this question, the Kruskal–Wallis test was run over the predicted results of different applied models (Table 12).

Table 12

The results of the Kruskal–Wallis test used to determine if there were statistically significant differences between the applied predictive models

Output parameterApplied modelsp-valueSignificantly different (95%)Significantly different (99%)
Lj RANFIS models 0.988 No No 
ANFIS models 0.999 No No 
ANFIS models, and RANFIS models No No 
ANFIS models, RANFIS models, and the empirical relation <0.0001 Yes Yes 
h2 ANFIS models, and the empirical relations <0.0001 Yes Yes 
ANFIS models No No 
Output parameterApplied modelsp-valueSignificantly different (95%)Significantly different (99%)
Lj RANFIS models 0.988 No No 
ANFIS models 0.999 No No 
ANFIS models, and RANFIS models No No 
ANFIS models, RANFIS models, and the empirical relation <0.0001 Yes Yes 
h2 ANFIS models, and the empirical relations <0.0001 Yes Yes 
ANFIS models No No 

Based on Table 12, in modeling the hydraulic jump at 95 and 99% confidence levels, no statistically significant difference exists between the results of ANFIS and RANFIS models. Furthermore, at these significance levels (0.05 and 0.01), no significant difference exists between the outputs of different algorithms. Findings on the secondary depth of jump (h2) and jump length (Lj) show that a significant difference exists between empirical relations (see Table 3) and soft computing models results at 95 and 99% confidence levels, while no significant difference is found between the applied soft computing models.

According to the summary results in Table 2 and the outcomes of this study, soft computing models are evidently superior to experimental relations; proportional to the results of the present study. Moreover, the superiority of the FA in combination with the recurrent ANFIS compared to the standard ANFIS is in line with the results obtained in this study.

CONCLUSIONS

Knowing hydraulic jump characteristics is a necessity for appropriate designing and planning of hydraulic structures such as stilling basins. In this study, the most important characteristics of the hydraulic jump (Lj and h2) on a rough bed were modeled and analyzed via ANFIS and RANFIS. In addition to ANFIS, five optimization algorithms (i.e., FA, PSO, GA, MFO, and WOA) were employed to train both ANFIS methods. To set up the ANFIS model, the input structure, including Ks, Q, h1, and B parameters, was used. Also, to establish the RANFIS structure, the h2 parameter as a recurrent factor was introduced to the model in addition to the mentioned inputs. The summary of the major findings of this study is presented below.

Results of h2 modeling indicated that soft computing models are more precise than experimental ones. The calculation of error measurement parameters showed that GA, PSO, and FA enhanced the efficiency of models. Nevertheless, MFO and WOA failed to improve the performance of models. Among these algorithms, ANFIS-FA had the best performance with RMSE, R2, and IA of 0.0053 m, 0.975, and 0.993, respectively. By considering RMSE as the basis, the use of GA, FA, and PSO improved the precision of ANFIS in modeling h2 by 30.26, 21.05, and 17.10%, respectively. According to the Lj predicted outcomes, even though the prediction efficiency of ANFIS and RANFIS methods are close, the application of recurrent architecture in the RANFISs enhanced the performance of ANFISs by around 3.50% (based on the RMSE values). However, to have a better view and judgment about the higher performance of recurrent neuro-fuzzy models, it is suggested that RANFIS methods be used and evaluated in other engineering phenomena. The comparison of the efficiency of the heuristic algorithms demonstrated that the FA and GA have a superior performance in combination with the RANFISs. The results also indicated that the FA, PSO, and GA improve the performance of the ANFISs and RANFISs. Thus, these three algorithms are suggested to be utilized as embedded learning techniques for integrative soft computing methods in future studies.

The results of sensitivity analysis for h2 input parameters showed that Q is more important than others in predicting h2 for weak, oscillatory, and steady jumps (1.7 < Fr < 9), while h1 is more important for strong jumps (Fr > 9). In terms of Lj, by increasing the Froude number (Fr > 9), the importance of h1 is increased and that of Q is decreased such that for a weak and oscillating jump, the importance of Q is higher than that of h1, while the opposite is true for the steady and strong jump. Results of the Kruskal–Wallis test revealed that, at 95 and 99%, a statistically significant difference exists between the intelligent and experimental method in modeling h2 (p < 0.0001), while no significant difference exists among soft computing models in modeling h2 and Lj.

COMPLIANCE WITH ETHICAL STANDARDS

The authors certify that they have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.

DATA AVAILABILITY STATEMENT

REFERENCES

Abbaspour
A.
Dalir
A. H.
Farsadizadeh
D.
Sadraddini
A. A.
2009
Effect of sinusoidal corrugated bed on hydraulic jump characteristics
.
Journal of Hydro-Environment Research
3
(
2
),
109
117
.
Abbaspour
A.
Farsadizadeh
D.
Ghorbani
M. A.
2013
Estimation of hydraulic jump on corrugated bed using artificial neural networks and genetic programming
.
Water Science and Engineering
6
(
2
),
189
198
.
Azimi
H.
Bonakdari
H.
Ebtehaj
I.
Michelson
D. G.
2018a
A combined adaptive neuro-fuzzy inference system–firefly algorithm model for predicting the roller length of a hydraulic jump on a rough channel bed
.
Neural Computing and Applications
29
(
6
),
249
258
.
Azimi
H.
Bonakdari
H.
Ebtehaj
I.
Gharabaghi
B.
Khoshbin
F.
2018b
Evolutionary design of generalized group method of data handling-type neural network for estimating the hydraulic jump roller length
.
Acta Mechanica
229
(
3
),
1197
1214
.
Azimi
H.
Bonakdari
H.
Ebtehaj
I.
2019
Gene expression programming-based approach for predicting the roller length of a hydraulic jump on a rough bed
.
ISH Journal of Hydraulic Engineering
,
1
11
.
doi:10.1080/09715010.2019.1579058
.
Azimi
A. H.
Shabanlou
S.
Yaghoubi
B.
2020
Prediction of hydraulic jump length on slope rough beds using extreme learning machine
.
Journal of Applied Research in Water and Wastewater
.
doi:10.22126/arww.2020.4427.1138
.
Bakhmeteff
B. A.
1932
Hydraulics of Open Channels
.
McGraw-Hill
,
New York
.
Bayon
A.
Valero
D.
García-Bartual
R.
López-Jiménez
P. A.
2016
Performance assessment of OpenFOAM and FLOW-3D in the numerical modeling of a low Reynolds number hydraulic jump
.
Environmental Modeling & Software
80
,
322
335
.
Carollo
F.
Ferro
V.
2004
Determinazione delle altezze coniugate del risalto libero su fondo liscio e scabro
.
Rivista di Ingegneria Agraria
4
,
1
11
.
Carollo
F. G.
Ferro
V.
Pampalone
V.
2007
Hydraulic jump on rough beds
.
Journal of Hydraulic Engineering
133
,
989
999
.
Cheong
D.
Kim
Y. M.
Byun
H. W.
Oh
K. J.
Kim
T. Y.
2017
Using genetic algorithm to support clustering-based portfolio optimization by investor information
.
Applied Soft Computing
61
,
593
602
.
Ead
S. A.
Rajaratnam
N.
2002
Hydraulic jumps on corrugated beds
.
Journal of Hydraulic Engineering
128
(
7
),
656
663
.
Gadbois
J.
Wilkerson
G.
2014
Uniform flow development length in a rough laboratory flume
. In:
World Environmental and Water Resources Congress 2014
,
Portland, OR
, pp.
1234
1242
.
Gerami Moghadam
R.
Yaghoubi
B.
Izadbakhsh
M. A.
Shabanlou
S.
2019
Prediction of the hydraulic jump length on sloping rough beds using meta-heuristic neuro-fuzzy model and differential evolution algorithm
.
Journal of Applied Research in Water and Wastewater
6
(
1
),
8
15
.
Govinda Rao
N. S. & Ramaprasad
1966
Application of momentum equation in the hydraulic jump
.
La Houille Blanche
4
,
451
453
.
Gu
S.
Bo
F.
Luo
M.
Kazemi
E.
Zhang
Y.
Wei
J.
2019
SPH simulation of hydraulic jump on corrugated riverbeds
.
Applied Sciences
9
(
3
),
436
.
Hager
W. H.
1992
Energy Dissipaters and Hydraulic Jump
.
Kluwer Academic Publisher
,
Dordrecht
, pp.
185
224
.
ISBN 0-7923-1508-1
.
Hager
W. H.
Bremen
R.
Kawagoshi
N.
1990
Classical hydraulic jump: length of roller
.
Journal of Hydraulic Research
28
(
5
),
591
608
.
Holland
J. H.
1975
Adaptation in Natural and Artificial Systems: an Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
.
U Michigan Press
,
Ann Arbor, MI
.
Houichi
L.
Dechemi
N.
Heddam
S.
Achour
B.
2013
An evaluation of ANN methods for estimating the lengths of hydraulic jumps in U-shaped channel
.
Journal of Hydroinformatics
15
(
1
),
147
154
.
Hughes
W. C.
Flack
J. E.
1984
Hydraulic jump properties over a rough bed
.
Journal of Hydraulic Engineering
110
(
12
),
1755
1771
.
Jang
J. S.
1993
ANFIS: adaptive-network-based fuzzy inference system
.
IEEE Transactions on Systems, Man, and Cybernetics
23
(
3
),
665
685
.
Karbasi
M.
Azamathulla
H. M.
2016
GEP to predict characteristics of a hydraulic jump over a rough bed
.
KSCE Journal of Civil Engineering
20
(
7
),
3006
3011
.
Kennedy
J.
Eberhart
R.
1995
Particle swarm optimization (PSO)
. In:
Proceedings of the IEEE International Conference on Neural Networks, Perth, Australia
, pp.
1942
1948
.
Leutheusser
H. J.
Kartha
V. C.
1972
Effects of inflow conditions on hydraulic jump
.
Journal of the Hydraulics Division
98
(
8
),
1367
1385
.
Li
X.
Sha
J.
Li
Y. M.
Wang
Z. L.
2018
Comparison of hybrid models for daily streamflow prediction in a forested basin
.
Journal of Hydroinformatics
20
(
1
),
191
205
.
Ma
F.
Hou
Y.
Prinos
P.
2001
Numerical calculation of submerged hydraulic jumps
.
Journal of Hydraulic Research
39
(
5
),
493
503
.
Mafarja
M.
Mirjalili
S.
2016
Whale optimization approaches for wrapper feature selection
.
Applied Soft Computing
62
,
441
453
.
Mahdavi-Meymand
A.
Scholz
M.
Zounemat-Kermani
M.
2019
Challenging soft computing optimization approaches in modeling complex hydraulic phenomenon of aeration process
.
ISH Journal of Hydraulic Engineering.
https://doi.org/10.1080/09715010.2019.1574619
McCorquodale
J. A.
1986
Hydraulic jumps and internal flows
. In:
Encyclopaedia of Fluid Mechanic
,
Vol. 2
,
Cheremisionoff
N.P.
Ed.,
Gulf Publishing
,
Houston
, pp.
120
173
.
Mirjalili
M.
Lewis
A.
2016
The whale optimization algorithm
.
Advances in Engineering Software
95
,
51
67
.
Mohamed Ali
H. S.
1991
Effect of roughened-bed stilling basin on length of rectangular hydraulic jump
.
Journal of Hydraulic Engineering
117
(
1
),
83
93
.
Montano
L.
Felder
S.
2020
LIDAR observations of free-surface time and length scales in hydraulic jumps
.
Journal of Hydraulic Engineering
146
(
4
),
04020007
.
Murali
P.
Revathy
R.
Balamurali
S.
Tayade
A. S.
2020
Integration of RNN with GARCH refined by whale optimization algorithm for yield forecasting: a hybrid machine learning approach
.
Journal of Ambient Intelligence and Humanized Computing
. https://doi.org/10.1007/s12652-020-01922-2.
Muzzammil
M.
Ayyub
M.
2010
ANFIS-based approach for scour depth prediction at piers in non-uniform sediments
.
Journal of Hydroinformatics
12
(
3
),
303
317
.
Naseri
M.
Othman
F.
2012
Determination of the length of hydraulic jumps using artificial neural networks
.
Advances in Engineering Software
48
,
27
31
.
Omid
M. H.
Omid
M.
Esmaeeli Varaki
M.
2005
Modeling hydraulic jumps with artificial neural networks
.
Proceedings of the Institution of Civil Engineers – Water Management
128
(
2
),
65
70
.
Pagliara
S.
Palermo
M.
2015
Hydraulic jumps on rough and smooth beds: aggregate approach for horizontal and adverse-sloped beds
.
Journal of Hydraulic Research
53
(
2
),
243
252
.
Pagliara
S.
Lotti
I.
Palermo
M.
2008
Hydraulic jump on rough bed of stream rehabilitation structures
.
Journal of Hydro-Environment Research
2
(
1
),
29
38
.
Peterka
A. J.
1958
Hydraulic Design of Stilling Basins and Energy Dissipaters
.
Engineering Monograph No. 25
,
U.S. Bureau of Reclamation
,
Denver, Colorado
.
Pourabdollah
N.
Heidarpour
M.
Koupai
J. A.
2019
An experimental and analytical study of a hydraulic jump over a rough Bed with an adverse slope and a positive step
.
Iranian Journal of Science and Technology, Transactions of Civil Engineering
43
(
3
),
551
561
.
Rajaee
T.
Mirbagheri
S. A.
Zounemat-Kermani
M.
Nourani
V.
2009
Daily suspended sediment concentration simulation using ANN and neuro-fuzzy models
.
Science of the Total Environment
407
(
17
),
4916
4927
.
Rajaratnam
N.
1967
Hydraulic jumps
.
Advances in Hydroscience
4
,
197
280
.
Rajaratnam
N.
1968
Hydraulic jumps on rough beds
.
Transactions of the Engineering Institute of Canada
11
(
A-2
),
1
8
.
Sharafati
A.
Tafarojnoruz
A.
Motta
D.
Mundher Yaseen
Y.
2020a
Application of nature-inspired optimization algorithms to ANFIS model to predict wave-induced scour depth around pipelines
.
Journal of Hydroinformatics
22
(
6
),
1425
1451
.
Sharafati
A.
Tafarojnoruz
A.
Shourian
M.
Yaseen
Z. M.
2020b
Simulation of the depth scouring downstream sluice gate: the validation of newly developed data-intelligent models
.
Journal of Hydro-Environment Research
29
,
20
30
.
Silvester
R.
1964
Hydraulic jump in all shapes of horizontal channels
.
Journal of the Hydraulics Division
90
(
1
),
23
56
.
Singh
K. M.
Sandham
N. D.
Williams
J. J. R.
2007
Numerical simulation of flow over a rough bed
.
Journal of Hydraulic Engineering
133
(
4
),
386
398
.
Taylor
K. E.
2001
Summarizing multiple aspects of model performance in a single diagram
.
Journal of Geophysical Research
106
,
7183
7192
.
Wei
X.
Zhang
L.
Yang
H. Q.
Zhang
L.
Yao
Y. P.
2020
Machine learning for pore-water pressure time-series prediction: application of recurrent neural networks
.
Geoscience Frontiers
12
(
1
),
453
467
.
Yang
X. S.
2010
Nature-Inspired Metaheuristic Algorithms
.
Luniver Press
,
Cambridge, UK
.
Zanganeh
M.
2020
Improvement of the ANFIS-based wave predictor models by the particle swarm optimization
.
Journal of Ocean Engineering and Science
5
(
1
),
84
99
.
Zhang
L.
Srisukkham
W.
Neoh
S. H.
Lim
C. P.
Pandit
D.
2018
Classifier ensemble reduction using a modified firefly algorithm: an empirical evaluation
.
Expert Systems with Applications
93
,
395
422
.
Zounemat-Kermani
M.
Scholz
M.
2013
Modeling of dissolved oxygen applying stepwise regression and a template-based fuzzy logic system
.
Journal of Environmental Engineering
140
(
1
),
69
76
.
Zounemat-Kermani
M.
Kisi
O.
Piri
J.
Mahdavi-Meymand
A.
2019a
Assessment of artificial intelligence-based models and metaheuristic algorithms in modeling evaporation
.
Journal of Hydrologic Engineering
24
(
10
),
04019033
.
Zounemat-Kermani
M.
Stephan
D.
Hinkelmann
R.
2019b
Multivariate NARX neural network in prediction gaseous emissions within the influent chamber of wastewater treatment plants
.
Atmospheric Pollution Research
10
(
6
),
1812
1822
.