A novel deep learning-based forecasting model optimized by heuristic algorithm for energy management of microgrid

•Development of data-driven approach to deal with uncertainties of RESs, load, and DAP for economic and robust operation of MG.•GA-AWPSO-LSTM-GAM model is designed to improve the forecasting accuracy in day-ahead prediction.•A detailed comparative analysis with respect to benchmark and state-of-art forecasting models is thoroughly conducted.•DM-CIDR program is adopted to consider specific attitudes exhibited by IDR customers according to their bid/offer data.•Results are obtained using a PJM market dataset for real applications.