Research On Platform Demand Response Modeling And Multi-Stage Optimization Strategy Under Cloud Edge Collaboration

With the double carbon target proposed,the energy structure of the power generation side is changing,and the installed capacity of renewable resources is increasing.However,the uncertainty of renewable resource generation has a great impact on the stable operation of the power grid.Demand response(DR)load and energy storage can effectively improve the operation performance of power grid by participating in system optimization and control.Based on the demand response load modeling,this paper establishes a multi-stage optimization model of distribution network combining demand-side load,energy storage and renewable resources under the cloud-edge collaboration framework,aiming to give an economical,effective and safe optimization scheme under the cloud-edge collaboration framework.Firstly,a typical three-level collaborative framework of distribution network cloud edge is introduced,and the functions of each part are introduced.The response principle of the power adjustment equipment(such as micro gas turbine,distributed energy storage unit,DR load)at the end of the distribution network is analyzed,and the global optimization of the day-ahead-day stage based on the cloud platform and the real-time stage coordinated optimization scheme of the cloud-edge collaboration are proposed.Secondly,a DR load model establishment strategy based on binary tree is proposed.The characteristics of residential electricity consumption are analyzed,and the DR load is classified according to the user ’s electricity preference.On this basis,the binary tree model representation method of DR load and the binary tree access architecture model are proposed.The binary tree aggregation model of DR load cluster regulation ability is established.Finally,the DR load cluster response ability of each stage is obtained.Finally,based on the proposed DR load cluster modeling method,a distribution network optimization strategy considering multi-time scale demand response is proposed by using the cloud edge collaboration framework.Using the global optimization function of the cloud center,in the day-ahead stage,the day-ahead DR load,the predicted intra-day DR load,the energy storage,and the electricity purchase and sale of the superior power grid are globally optimized.In the intra-day stage,according to the multi-stage characteristics,an intra-day rolling optimization model is proposed to achieve the purpose of approaching the day-ahead response scheme.In the real-time stage,the fully distributed consensus algorithm is used to optimize the interaction power of each edge cloud,so as to achieve the purpose of matching the load curve with the lowest compensation cost.Through the simulation of a regional distribution network example,it is verified that the proposed multi-stage collaborative optimization strategy under cloud-edge collaboration can effectively reduce the control cost and improve the real-time performance of optimization.