Deep learning based real time Demand Side Management controller for smart building integrated with renewable energy and Energy Storage System

Electric Power Consumption (EPC) and Renewable Energy Generation (REG) are inconsistent and this fickle nature affects the utility grid's power quality and system stability. The electric transmission and distribution infrastructure must be upgraded to meet the consumer's peak demand. Hence, proposing a Demand Side Management (DSM) program in smart grid to reduce utility grids Peak to Average Ratio (PAR) and end-users electricity tariff. Renewable energy with Energy Storage System (ESS) in the DSM controller is used to enhance the end user's economic and environmental features. This article proposes a Recurrent Neural Network (RNN) based Long Short Term Memory (LSTM) framework for Science Block (SCB) every minute and 5 min of EPC and REG forecasting to develop the DSM program. This proposed deep learning model performance, is evaluated using Mean Squared Error (MSE), Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and R-squared. The results show that the proposed DSM program benefits end electricity users and smart grid operators. [Display omitted] •The Demand Side Management (DSM) controller is analyzed in a real-time working environment.•Short-term forecasting of Energy Consumption (EC) and Renewable Energy Generation (REG) is done using the LSTM model.•The DSM controller generates control signal based on the dynamic electricity price, forecasted EC and REG.•This proposed DSM controller reduces the utility grid peak demand and end-user electricity tariffs.•In the proposed system, energy storage plays a most significant role in renewable energy-integrated smart grid systems.