A Multi-Period Joint Energy Scheduling Algorithm In Smart Home Based On Prediction Of The Residents Energy Consumption

In this paper, considering a tradeoff between consumers comfort and energy efficiency, a multiperiod joint energy scheduling algorithm (MPJ-ESA) based on prediction of residents energy consumption is proposed, which includes long-period preliminary sch eduling, short-period preliminary scheduling, and real-time fine-tuning scheduling. First, by analyzing historical data of energy consumption, preferred usage profile of consumers is inferred, and the dynamic comfort level is presented. Then the paper uses the wavelet neural networks (WNNs) prediction algorithm to predict the operation of the appliances which are classified into appliances with unschedulable mode and schedulable mode. Based on the energy consumption prediction and dynamic comfort level, home appliances running state are scheduled according to the prediction of renewable energy available amount and real-time pricing (RTP). The simulation results show that scheduling algorithm effectively improves the energy efficiency and enhances user satisfaction with the operation of scheduled appliances and let the consumers comfort and energy efficiency achieve a better tradeoff.