neural-fuzzy inference system

The use of labyrinth weirs is one of the most effective and economical ways to increase weir efficiency by enhancing weir crown length, so that at a specified width and hydraulic height compared to other weirs, more discharge passes. In this study, the hybrid method of the Whale optimization algorithm and the fuzzy neural- inference system to estimate the optimal discharge coefficient of labyrinth weirs with the curved plan is introduced. In the proposed method, the efficiency and convergence of the Whale algorithm have been measured. The experimental data were used to validate the proposed method. The performance of the proposed method was evaluated with four statistics, including the correlation coefficient (R2), the root means square error (RMSE), the mean absolute error (MAE) and the efficiency criterion (NSE). The results indicate a good match between the observed and estimated values. The results of the WOA-ANFIS method with (H /P) and (θ), showed that the discharge coefficient of labyrinth weirs with the curved plan (Cd=0.786) in compared to labyrinth weirs with a triangular plan (Cd=0.682), it increases by 10%. Also, statistical analyzes performed on the obtained results showed that the WOA-ANFIS method with high values of RMSE = 0.021, R2 = 0.981, MAE = 0.010 and NSE = 0.976 has high efficiency in estimating the discharge coefficient of this type of weirs.