Research On "Source-Network-Load" Coordination Planning Model For Active Distribution Network

In order to consume more distributed power sources and facilitate the low-carbonization of terminal power,and to ensure the safety and reliability of the distribution network operation through more flexible control methods,the traditional distribution network has made a transition to the active distribution network.With the coordinated development of energy storage technology and the demand side response,more load becomes available resources.Active distribution networks need to consider more uncertain factors of source layer,network layer and reservoir layer,so they should be considered in planning.The decision variables are broader,more diverse,and more diverse.On the other hand,in the gradual liberalization of the power market,there are more and more independent entities,such as distributed power operators,distribution network companies,and user-side demand response agents,which are needed for active distribution network planning.Considering the possible multi-agent independent decision-making but affecting the actual situation,studying the multi-party main game relationship and constructing the multi-agent interest coordination planning model are also the research directions to be considered for the active distribution network planning.This thesis first establishes a multi-objective bi-level planning model for active distribution network considering economy,reliability and stability.The main decision-making content is the distributed power supply location and grid upgrade.The upper layer of the model considers the planning plan from the investment level,and the lower level operation level considers the active management measures.An improved multi-objective particle swarm optimization algorithm is used to solve the model.The result of the model solution can obtain a distributed power supply location and fixed volume investment decision and a new layout plan for network upgrade with consideration of economy,reliability and stability.Then,based on the above-mentioned distributed power supply capacity selection and grid coordination planning,this thesis considers the improvement of DG output and the response of demand response to load timing characteristics of energy storage charge and discharge,and considers energy storage and demand side management.Active distribution network planning model.Through simulation examples,considering the energy storage and discharge factors can smooth the DG output curve,slow down the node power impact,improve the DG utilization rate,reduce the amount of abandoned wind and reduce the distributed power investment cost,and consider the demand side management can reduce the power supply of the main network.The burden,delaying the investment in upgrading the distribution network,has improved the economics of planning investment and the security and stability of system operation as a whole.Finally,for the multi-party independent subject game relationship existing in the power market,the distributed power supply operator,the distribution network company,and the user-side demand response agent are the three parties.Considering the three parties’ pursuit of maximizing their respective interests,the decision is made.Impact and constraints.Based on the above-mentioned planning model,the influence of the access and demand response of distributed power supply on the uncertainty of active distribution network planning is considered,and the multi-agent interest game coordination planning model of active distribution network “source-network-load” is established.The model seeks a balanced solution with high interest among all parties through the game relationship.The solution plan solves the problem of improving the capacity of distributed power supply,improves the revenue of distributed power operators,reduces the planning cost of distribution companies and improves overall revenue.The enthusiasm of the load demand responds to the peak load and valley filling effect of the overall load,and achieves the purpose of promoting “sourcenetwork-load” multi-agent friendly interaction and coordinated planning.