JWSS - Journal of Water and Soil Science

In this study, the groundwater level (GWL) of the Sarab Qanbar region located in the south of Kermanshah, Iran, was estimated using the Wavelet- Self- Adaptive Extreme Learning Machine (WA- SAELM) model. An artificial intelligence method called “Self- Adaptive Extreme Learning Machine” and the “Wavelet transform” method were implemented for developing the numerical model. First, by using the autocorrelation function (ACF), the partial autocorrelation function (PACF) and the effective lags in estimating GWL, eight distinctive SAELM and WA- SAELM models were developed. Later, the values of the observational well were normalized for estimating GWL. Next, the most optimized mother wavelet was chosen for the modeling. By evaluating the results of SAELM and WA- SAELM, it was concluded that the WA- SAELM models could estimate the values of the objective function with higher accuracy. Then, the superior model was introduced, showing that it could be very accurate in forecasting the GWL. In the test mode, for example, the values of R (correlation coefficient), Main absolute error (MAE) and the NSC- Sutcliffe efficiency coefficient (NSC) for the superior model were calculated to be 0.995, 0.988 and 0.990, respectively. Furthermore, an uncertainty analysis was conducted for the numerical models, proving that the superior model had an underestimated performance.

The river flow prediction is a key aspect of hydrology that plays a significant role in water resources management, flood risk reduction, and agricultural planning. This study simulates the monthly flow of the Razavar River, located in western Iran, using an extreme learning machine (ELM) model enhanced by the Whale (WOA) Optimization Algorithm and Grasshopper Optimization Algorithm (GOA) metaheuristic optimization algorithms. The data used include river flow, precipitation, evaporation, and temperature, which were collected for 10 years with a monthly time step and normalized in the numerical range of zero to one. 80% of the data is used for training, and the remaining 20% for model evaluation. The performance of the models is measured with the statistical indices RMSE, NSE, and R². First, the basic ELM model is developed using the trial-and-error method to adjust the weights between the hidden and output layers. Then, the WOA and GOA algorithms are used to optimize the weights. The results show that the basic ELM model performs worse than the optimized models (Train: RMSE=0.1427, NSE=0.7795, R²=0.7911, Test: RMSE=0.1406, NSE=0.7811, R2=0.7916). While the WOA-ELM and GOA-ELM models provide similar results, the WOA-ELM model shows better performance in complex conditions (Train: RMSE=0.1215, NSE=0.7869, R2=0.7932, Test: RMSE=0.1165, NSE=0.7872, R2=0.7933). The results of this research show that meta-heuristic optimization algorithms play an important role in improving the performance of river flow prediction models due to their ability to search comprehensively and avoid getting stuck in local optima. The findings of this study emphasize the importance of applying these techniques in water resources management and sustainable planning and will pave the way for future research in this area.