SHORT-TERM RESIDENTIAL LOAD FORECASTING USING DEEP LEARNING TECHNIQUES

Abstract

Smart energy management systems have become more popular as the consumption of energy is increasing rapidly. So, in order to monitor the daily energy consumptions in real time smart meters are used. Load forecasting is very important for power system management as it helps in maximum utilization of power generation plants, reliable and efficient operation of the system. In smart homes, the smart meter data are used to forecast the load and it can be even used for a neighborhood. F orecasting of electrical loads can be done using different deep learning techniques and can be used for demand management. Different methods employed includes Long Short-Term Memory (LSTM), Bi-directional Long Short-Term Memory (Bi-LSTM) and Gated Recurrent Unit (GRU). The proposed method for forecasting of energy consumption consists of data preprocessing, model generation and validation. Performance of the models are validated using evaluation metrics like R-squared (R2 ), Mean Absolute Error (MAE), Mean Square Error (MSE) and Root Mean Square Error (RMSE). The error metrics are then compared to find out the accurate model. The main advantage of load forecasting is that we can reduce the energy wastage and increase the efficiency of energy usage