This study investigates the performance of artificial intelligence techniques including artificial neural network (ANN), group method of data handling (GMDH) and support vector machine (SVM) for predicting water quality components of Tireh River located in the southwest of Iran. To develop the ANN and SVM, different types of transfer and kernel functions were tested, respectively. Reviewing the results of ANN and SVM indicated that both models have suitable performance for predicting water quality components. During the process of development of ANN and SVM, it was found that tansig and RBF as transfer and kernel functions have the best performance among the tested functions. Comparison of outcomes of GMDH model with other applied models shows that although this model has acceptable performance for predicting the components of water quality, its accuracy is slightly less than ANN and SVM. The evaluation of the accuracy of the applied models according to the error indexes declared that SVM was the most accurate model. Examining the results of the models showed that all of them had some over-estimation properties. By evaluating the results of the models based on the DDR index, it was found that the lowest DDR value was related to the performance of the SVM model.

Side weirs have many possible applications in hydraulic engineering field. They are also considered as an important structure in hydro systems. In this study, support vector machine (SVM) technique was employed to predict the side weir discharge coefficient. The performance of SVM was compared with other types of soft computing techniques such as artificial neural network (ANN) and adaptive neuro fuzzy inference systems (ANFIS). While ANN and ANFIS models provided a good prediction performance, the SVM model with a radial basis function kernel function outperforms them. The best SVM model was developed with a gamma coefficient and epsilon of 15 and 0.3, respectively. The SVM yielded coefficient of determination (R²) equal to 0.96 and 0.93 for the training and testing data. Sensitivity analysis of the ANN, ANFIS and SVM models showed that the Froude number and ratio of weir length to the flow depth at upstream of weir are the most effective parameters for the prediction of discharge coefficient.

Modeling of river phonemes is one of the main parts in Hydraulic engineering studies. Predicting the discharge in the compound open channel is the main subject of the river engineer researches. In this study the support vector machine (SVM) technique is used to predict the discharge in the compound open channel. The result of the error indexes shows that the best accuracy of the SVM by correlation coefficient 0.99 is related to the SVM model which includes the radial basis function (RBF) as kernel function. During the SVM model process obtained 0.00005 for the gamma coefficient, epsilon is equal to 0.1 and C = 1.

Flow pattern recognition using the physical modeling and numerical simulation helps designers to propose optimal shape for hydraulics structures. Optimal shape of hydraulic structure causes increasing their performance. In this study, the effect of geometries of guide walls on the flow pattern and rating curve of Balaroud dam spillway’s (Iran) was simulated using numerical and physical simulation. Numerical simulation of flow pattern through the guide walls was carried out using Flow 3D software as computational fluid dynamic (CFD) tool. To this task, three plans for the guide walls were evaluated. The results showed that the plan (3) has best performance for removing cross waves and smoothly passing the flow through the approach channel and over the spillway. Evaluating the turbulence models showed that the RNG-K-epsilon becomes the best performance to show the cross waves

Modeling of river pollution contributes to better management of water quality and this will lead to the improvement of human health. The advection dispersion equation (ADE) is the government equation on pollutant transmission in the river. Modeling the pollution transmission includes numerical solution of the ADE and estimating the longitudinal dispersion coefficient (LDC). In this paper, a novel approach is proposed for numerical modeling of the pollution transmission in rivers. It is related to use both finite volume method as numerical method and artificial neural network (ANN) as soft computing technique together in simulation. In this approach, the result of the ANN for predicting the LDC was considered as input parameter for the numerical solution of the ADE. To validate the model performance in real engineering problems, the pollutant transmission in Severn River has been simulated. Comparison of the final model results with measured data of the Severn River showed that the model has good performance. Predicting the LDC by ANN model significantly improved the accuracy of computer simulation of the pollution transmission in river.