A novel IBAS-ELM model for prediction of water levels in front of pumping stations

•The search ability of the BAS algorithm is improved by increasing the search direction and introducing the optimal individual and levy flight.•The IBAS algorithm is used to optimize the weight and bias of ELM to improve the prediction accuracy.•The data are divided into regulated and unregulated periods for water level prediction according to whether the unit is regulated or not.•Model input in the regulated period is improved considering the regulation time and the discharge of the pumping station.