Research Of Multi Time-scale Scheduling Strategy For Smart Residential Electricity Consumption

Electricity consumption is the final link of power system transmission and distribution,also the final goal of power transmission.Residential electricity consumption is an important part of the consumption link,and accounts for a large proportion in the total electricity consumption of the society.Currently the residential electricity consumption has the characteristics of large cardinal and low intelligent level.All pieces of electrical equipment in the residential consumer are separately controlled,as a result there is lack of united control and optimization for residential electrical equipment.With the popularization and application of family distributed photovoltaic(PV)system and energy storage system,residential loads will be transformed from the pure loads to the microgrids with energy source,energy storage,loads and so on.With the application of home energy management system(HEMS)the electrical equipment in the residential consumer can be coordinately controlled and the residential electricity consumption can be overall optimized,so that the cost of electricity consumption can be reduced and the local consumption of PV can be improved.Considering that the output power of PV system and the consumption behavior of residential consumers both have uncertainty,and the forecast of them have different accuracy in different time-scales,the multi time-scale scheduling strategy for smart residential electricity consumption is proposed.In the day-ahead scheduling stage,for residential consumers with PV system and energy storage system,in view of the characteristics that the forecast accuracy of PV systems and consumption behaviors of residential consumers decrease gradually with the growing time-scale,the models for distributed energy sources,residential loads and the uncertainty of residential consumers’ behavior are built.A stochastic and adjustable robust optimization hybrid co-scheduling strategy for smart residential electricity consumption is proposed,taking advantages of both optimization methods.It is aimed at minimizing the operating cost of the system and guaranteeing the comfort of consumers and the electricity consumption freedom.With the engineering game theory and Improved Particle Swarm Optimization,the optimization model is transformed to Mixed Integer Linear Programming problem.In the intra-day scheduling stage,based on the day-ahead scheduling plan,the intra-day scheduling strategy of residential consumers based on Model Predictive Control(MPC)and multi-scenarios stochastic optimization is proposed,considering the requirements and characteristics of intra-day scheduling.The operation plans of schedulable and uninterruptable loads are the same as those from day-ahead scheduling plan.The strategy makes rolling optimization with MPC in the day with smaller time-scale,and takes the operation plan of energy storage system from the day-ahead scheduling plan as reference to build constraints.Through the simulation on the test system built in the paper,the effectiveness of the proposed multi time-scale scheduling strategy for smart residential electricity consumption is proved,including in reducing the cost of the consumers,improving the local consumption of PV,following operation status of the system in the day,adjusting the control strategies and so on.