Search

The sedimentation in coastlines, waterways, and reservoirs is an environmental problem that can be coped with sediment removal systems such as hydro-suction. In this study, the Mayfly Algorithm (MA) is applied to predict the maximum depth (hs) and diameter (Ds) of the formed scour hole attributed to the hydro-suction. Two machine learning methods including ANFIS and SVR were trained by the MA and Genetic Algorithm (GA) based on the measured data from this research as well as gathered data from pertinent resources. The most accurate methods in predicting hs and Ds were ANFIS-MA and SVR-MA, respectively. The results showed that the MA improved the efficiency of ANFIS and SVR about 23.69% and 13.56%, whereas the performance enhancement for GA were about 17.06% and 7.98%, respectively. Comparing the results between the GA and MA demonstrated that the developed MA-embedded integrative methods are more accurate than the GA-embedded methods. [ABSTRACT FROM AUTHOR]

Reliable and accurate modeling of groundwater quality is an important element of sustainable groundwater management of productive aquifers. In this research, specific conductance (SC) of groundwater is predicted based on different individual and integrative machine learning, adaptive neuro‐fuzzy inference system (ANFIS), and nonlinear mathematical models. For developing the integrative models, the well‐known particle swarm optimization (PSO) and novel manta ray foraging optimization (MRFO) heuristic algorithms are embedded in the models. Presenting different univariate, bivariate, and multivariate input scenarios, the parameters used to develop and validate the models include groundwater level, salinity, and water temperature at an observation well near Florida City. The findings reveal that applying more independent parameters (multivariate scenario) enhances the performance of both the mathematical and machine learning models. Even though the mathematical models present an acceptable performance for the prediction of SC (index of agreement, IA, equals 0.933), the ANFIS models provide the most accurate SC predictions (IA = 0.943). Both the PSO and MRFO algorithms improved the prediction capability of the ANFIS models with, respectively, 13% and 5% for the RMSE. [ABSTRACT FROM AUTHOR]

The present study investigates the capability of two metaheuristic optimization approaches, namely the Butterfly Optimization Algorithm (BOA) and the Genetic Algorithm (GA), integrated with machine learning models in Suspended Sediment Load (SSL) prediction. The Adaptive Neuro-Fuzzy Inference System (ANFIS), Artificial Neural Network (ANN), and Multiple Linear Regression (MLR) are applied as the predictive data-driven models. Independent input variables, i.e., the water temperature (T), river discharge (Q), and specific conductance (SC) are used for the prediction of SSL based on several statistical indices. The results indicate that the performances of all studied models were close to one another; moreover, the metaheuristic algorithms were found to increase the accuracy of the ANFIS and ANN models for approximately 11.73 percent and 4.30 percent, respectively. In general, the BOA outperformed the GA in enhancing the optimization performance of the learning process in the applied machine learning models. [ABSTRACT FROM AUTHOR]

In this study, multi-tracker optimization algorithm (MTOA), particle swarm optimization (PSO), and differential evolution (DE) algorithms were integrated with support vector regression (SVR) to predict energy dissipation downstream of labyrinth weirs (Δ E). In order to evaluate the performance of these methods, the results are compared with corresponding outcome obtained by applying two other methods, namely, multilayer perceptron neural network (MLPNN) and multiple linear regressions methods (MLR). The input parameters comprise the discharge, the upstream flow depth, the crest length of a single cycle of the labyrinth weir, the width of a single cycle of the labyrinth weir, the apex width, the number of labyrinth weir cycles, the sidewall angle, and the height of weir. The results indicate that the meta-heuristic algorithms substantially improve the performance of SVR. The results show that the integrative methods, SVR-MTOA, SVR-PSO, and SVR-DE, are more accurate than the MLPNN and the MLR. In average, the integrative methods provide 39.63% more accurate results than the MLPNN and 79.34% more accurate results than the MLR. The average RMSE and R2 for the integrative methods are 0.0054 m and 0.977, respectively. Among all integrative methods, the SVR-MTOA yields the best results, with RMSE = 0.0044 m and R2 = 0.986. [ABSTRACT FROM AUTHOR]

This study investigates and challenges the capability of standard and hybrid soft computing models of fuzzy c-means clustering adaptive neuro-fuzzy inference system (ANFIS), wavenet and artificial neural networks (MLPNN and RBFNN) to estimate the spillway aerator air demand in dams. For the learning process, four different meta-heuristic optimization methods (particle swarm optimization (PSO), genetic algorithm (GA), firefly algorithm (FA) and biogeography-based optimization (BBO)) are considered as alternatives to the classical optimization algorithms of the data-driven models. In addition to the data-driven models, the multiple linear regressions and some empirical relations are used to evaluate the performance of the models. Evaluation of the models is assessed with five different statistical parameters as well as the diagnostic tool of the Taylor's diagram. Analysis of the models' outcome reveals that the ANFIS-GA has the best performance associated with a standard root mean square error of 0.309 and a coefficient of determination (R2) of 0.93. [ABSTRACT FROM AUTHOR]