Efficient residential load forecasting using deep learning approach
Rida Mubashar, Mazhar Javed Awan, Muhammad Ahsan, Awais Yasin, Vishwa Pratap Singh
Abstract
Reliable and efficient working of smart grids depends on smart meters that are used for tracking electricity usage and provides' accurate, granular information that can be used for forecasting power loads. Residential load forecasting is indispensable since smart meters can now be deployed at the residential level for collecting historical data consumption of residents. The proposed method is tested and validated through available real-world data sets. A comparison of LSTM is then made with two traditionally available techniques, ARIMA and Exponential Smoothing. Real data from 12 houses over a period of 3 months is used to inspect and validate the accuracy of load forecasts performed using three mentioned techniques. LSTM models, due to their higher capability of memorising large data, establish their utilisation in time series-based predictions.