Application of soft computing and evolutionary algorithms to estimate hydropower potential in multi-purpose reservoirs

Abstract Hydropower is a clean and efficient technology for producing renewable energy. Assessment forecasting of hydropower production are important strategic decision-making. This study aimed to use machine learning models, including adaptive neuro-fuzzy inference system (ANFIS), gene expression programming, random forest (RF), least square support vector regression (LSSVR), predicting hydroelectric energy production. A total eight input scenarios was defined with combination various observed variables, evaporation, precipitation, inflow, outflow the reservoir, predict produced during experimental period. The Mahabad reservoir near Lake Urmia in northwest Iran selected as object. results showed that previous month, from dam resulted highest prediction performance using RF model. scenario included all variables except precipitation outperformed other LSSVR Among exerted which RMSE, MAPE, NSE were 442.7 (MWH), 328.3 0.85, respectively. Harris hawks optimization (HHO) (RMSE = 0.2 WMH, MAPE 10 0.90) better than particle swarm (PSO) optimizing ANFIS prediction. Taylor’s diagram indicated ANFIS-HHO model had accuracy. findings this models can be used an essential tool decision-making sustainable