Distributed Wind Power And Photovoltaic Planning Based On An Improved Clustering-Scenario Analysis Method

In the modern era,the state advocates renewable energy instead of fossil fuels,and encourages more and more renewable energy access to the power grid,which can be connected to the power grid in the form of distributed generation.Distributed generation is now an important measure of national energy transformation,so the planning of distributed generation is a hot topic.Current distributed generation planning is static planning,regardless of the temporal characteristics randomness and of distributed generation output,regardless of their reactive power regulation characteristic at the same time,so this conservative planning results can not promote the access of distributed generation.Therefore,considering the randomness of the distributed power output and load and distributed generation regulation characteristics,exploring the access position,capacity and type of distributed generation has certain theoretical meaning and engineering value.From the perspective of independent power generation,the mathematical model of the maximum annual income of distributed generation is established.In combination with the actual situation,the calculation of reactive power regulation of distributed generation is added.The algorithm of distributed generation planning is constructed by using the memetic algorithm.According to the planning problem,this paper analyzed the temporal characteristics of distributed generation output and load,established original typical timing scenario set to deal with the randomness,using affinity propagation clustering algorithm to scene cuts,to improve the efficiency of algorithm.In order to avoid the interference of clustering results in the actual data,the correntropy function is introduced to improve the clustering algorithm.Using IEEE33 node system and the actual engineering example simulation tests the validity of the programming model and algorithm.Compared to traditional planning and typical day planning,it proved in this paper that scenario analysis can effectively consider the randomness of the distributed generation and promote to adopt the new energy.On the basis of scene analysis,a multi-objective distributed generation planning model is established from the perspective of different interests,and the multi-objective memetic algorithm is used to solve the problem.Simulation tests on IEEE33 nodes demonstrate the effectiveness of the proposed model and multi-objective algorithm,which can provide decision makers with the optimal configuration scheme from different interest entities perspectives.