Research On Multi-time Scale Scheduling Of Integrated Energy System With Wind And Solar Storage

With the gradual depletion of traditional fossil energy sources,clean and renewable energy sources have attracted people’s attention,and the installed capacity of new energy power generation has also increased year by year.In order to fully absorb new energy and improve the energy utilization ratio of the system,the concept of Integrated Energy Systems(IES)came into being.Compared with the traditional single-stage energy utilization,IES realizes the multi-stage utilization of cold,heat,electricity,gas and other forms of energy.However,due to the large number of devices,the variety of energy flow forms,and the coupling relationship between different energy sources in the IES,it leads to the complexity of energy optimal scheduling.It needs a reasonable optimal scheduling method to distribute energy,ensure it can operate safely and stably.In this paper,the research work on multi-time scale dispatching of IES mainly includes three parts.In the first part,aiming at the problem that the traditional wind,photovoltaic and load power forecasting methods can’t meet the actual optimal scheduling accuracy requirements,the BP neural network based on the self-learning teaching and learning optimization algorithm is used to predict the wind,photovoltaic and load power,and a better prediction effect is obtained compared with the traditional algorithm.In the second part,with the goal of economy and environmental protection,a 0-1 mixed integer nonlinear programming multi-objective programming model with time-sharing pricing strategy is builded,and simulation studies are carried out in summer,winter and transitional season.The results show that the optimal day-ahead scheduling can be achieved for different seasons by using the Opti toolbox to solve the above models.In the third part,in order to solve the problem that the day-ahead scheduling plan can’t be directly applied to the IES,this paper applies the Model Predict Control(MPC)to the real-time rolling scheduling stage,and observe the change of system output by applying disturbance.The simulation results show that the MPC strategy can track the day ahead scheduling plan better and has better anti-jamming capability.