A Data Driven Approach for Day Ahead Short Term Load Forecasting

This paper aims to develop an evolutionary deep learning based hybrid data driven approach for short term load forecasting (STLF) in the context of Bangladesh. With lapse time, power system is getting complex. Penetration intermittent renewable energy (RE) into grid, changing prosumer pattern with need demand side management (DSM) has thrown a challenge dynamic operation and control. Load plays significant role this In addition, it directly affects future planning network expansion, unit commitment economic mix market. Day ahead short-term very crucial day operation. As such, various conventional modified methods have been used over time prediction. Nevertheless, existing approaches like age old statistical methods, artificial intelligence (AI), machine (ML), (DL) techniques alone cannot provide effective accuracy all time. Hence, integrated genetic algorithm (GA)-bidirectional gated recurrent (Bi-GRU) technique (GA-BiGRU) proposed work. The developed method validated Bangladesh (BPS) by providing electrical whole country. Besides, performance prediction model compared some such as long memory (LSTM), (GRU) algorithm-gated (GA-GRU) terms mean absolute error (MAPE) root squared (RMSE). outcome gives indication better GA-BiGRU DL others.