Day-ahead Hierarchical And Partitioned Coordinated Optimal Scheduling For Active Distribution Network By Considering Interests Of Multi-stakeholders

With the continuous expansion of active distribution networks(ADN)and the increase of the penetration of renewable energy sources(RES),the initiative of the sources,networks and loads in ANDs is greatly enhanced,which leads to the sharp increase of control variables and complexity in distribution network(DN)optimization.In addition,the stakeholders involved in ADN optimization are more diversified and the privacy protection for different stakeholders is becoming more important.These factors bring great challenges to the optimal scheduling of ADNs.The hierarchical and partitioned coordinated optimal scheduling is a potentially efficeient way to solve these challeges.Therefore,the research on the day-ahead hierarchical and partitioned coordinated optimal scheduling of the ADN is developed in this paper.The main contributions are as follows:Firstly,the active distribution network is divided into three layers structually:the DN layer,the microgrid(MG)layer and the User layer.Two kinds of day-ahead three-layer coordinated optimization scheduling mechanisms are proposed,which consider the benefits of the DN,the MGs and the users at the same time.For the two mechanisms,the optimization models are established for the DN layer,the MG layer and the user layer,respectively.In the user layer,the user optimization model is established and then the scheme of user loads and battery outputs is formulated by solving the model with the seqential quantitative programming(SQP);In the MG layer,the demand response strategy based on electricity price is proposed and then the MG-user joint optimization solved by the differential evolution(DE)algorithm is carried out with considering the user layer optimization results as constraints;In the DN layer,the DN adjusts the reactive power compensation values to optimize their own interests and ensure the stable operation of the whole system and then the interior point method(IPM)is used to solve the DN layer optimization problem.Based on a 10 kV distribution system with 47 nodes in Panzhihua City,two simulation tests are conducted to verify the effectiveness of the proposed three-layer day ahead coordinated optimization methodsSecondly,the partition principle of the DN considering the efficiency of multi-area parallel optimization and the safe operation of DN is proposed.According to the partition principle,the multi-objective optimization partition model is established and solved by NSGA-Ⅲ algorithm.The influence of DN partition on the efficiency of multi-area coordinated optimization is analyzed and three partition principles,i.e.,reactive voltage control partitioning,user quantity balancing partitioning and source-load power balancing partitioning,are proposed.In the reactive voltage control partitioning,the electric distance and Fast-Newman theory are applied and then maximizing modular function Q is set to the optimization objective.In the user quantity balancing optimization partitioning,the objective is to minimize the mean square deviation of the user quantities in all partitions.In the source-load balancing optimization partitioning,the optimization objective is to minimize the sum of the maximum source-load unbalanced power in all partitions.According to the structure of DN,the connectivity constraint and partition size constraint are proposed and then the constraints are solved by improving the population initialization method and mutation method in NSGA-Ⅲ.The method is verified by a 10kV distribution system with 47 nodes,124 nodes and 300 nodes respectively in Panzhihua City.The results show that the method has a strong guiding significance for DN partitioning.Finally,based on the comprehensive application of ADN hierarchical partitioned coordination optimization scheduling and DN optimization partitioning method proposed in this paper,a day-ahead coordinated optimization scheduling method for multi inter-connected MGs is proposed.In the method,the power flow constrains are considered in each MG A parallel iterative method is proposed to solve the coordinated optimization of multi inter-connected MGs system.Based on MATLAB,a multi-computer parallel computing platform is built.Simulation cases are designed based on a 10kV distribution system with 47 nodes,124 nodes and 300 nodes respectively in Panzhihua City.The optimization results of different partition schemes are compared to verify the effectiveness of the proposed method.