Forecasting of Electric Load Using a Hybrid LSTM-Neural Prophet Model

Load forecasting (LF) is an essential factor in power system management. LF helps the utility maximize utilization of power-generating plants and schedule them both reliably economically. In this paper, a novel hybrid method proposed, combining long short-term memory network (LSTM) neural prophet (NP) through artificial network. The paper aims to predict electric load for different time horizons with improved accuracy as well consistency. proposed model uses historical data, weather statistical features obtained from data. Multiple case studies have been conducted two real-time data sets on three types forecasting. later compared few established methods found literature performance metrics: mean average percentage error (MAPE), root square (RMSE), sum (SSE), regression coefficient (R). Moreover, guideline various attributes provided considering applications model. results comparisons our test cases showed that over other techniques.