Demand forecasting in smart homes using deep learning techniques

As energy usage grows quickly, smart energy management solutions have gained in popularity. Smart meters are used to track daily energy consumption in realtime. Demand forecasting is very important in smart homes where the smart meter data is used to forecast the load of a particular household and it can be even used for a neighbourhood. Prediction of electrical loads can be done using different deep learning techniques and can be used for demand management. Deep learning techniques proposed in this work include LSTM, Bi-LSTM and GRU. The proposed model for the prediction of energy consumption consists of data preprocessing, model generation and validation. The main benefit of this demand forecasting is to decrease energy wastage and improve energy consumption effectiveness.