Domestic Energy Consumption Forecasting using Machine Learning

Energy management is essential maintain the demand at consumer end, reduce the wastage of electricity during off peak hours, environmental constrains, and large transmission and distribution losses. Consumers enabled with renewable energy sources and electric vehicles are the key for the intelligent and green grid. Renewable energy sources installed near load centres to manage the demand and electric vehicles are used as backup source of energy and supply to grid during peak hours. Demand of consumers is random and to maintain the demand with renewable energy sources and electric vehicles, connected appliances at consumer end are categorized as shiftable, non-shiftable and curatable appliances. To manage the demand for a reliable grid, energy consumption prediction is essential at consumer end for the demand. To. forecast the demand a long short term memory algorithm is used and evaluate the error's obtained in the forecasting. In future this model can integrate with game theory for energy bidding.