Application Research On Multi-objective Optimization Of Power Selling Packages Based On Soft Set Theory

The 9th power reform documen confirms the orientation of the reform of the power market in China,that is introducing competition in the power selling link,which mainly includes opening up the users' free choice,allowing users to choose the power selling companies freely,and finally forming a competitive market pattern for many power selling companies.In the future market will emerge a variety of power selling packages,ordinary users in the face of a variety of power selling packages will face a big problem: how to choose their own affordable power selling packages.Therefore,the application research of multi-objective optimization of power selling packages is of great significance to the reform of power selling side in power market.Firstly,this paper introduces the current situation and development trend of domestic and foreign electricity sales side reform.Based on the analysis of overseas power selling packages and the actual situation of China's electricity market,the power selling packages in line with China's electricity market are put forward.The application status and development trend of soft set theory and clustering analysis algorithm are analyzed,and the application of soft set theory and clustering analysis algorithm to multi-objective optimization of power selling packages is proposed.Secondly,the K-means clustering algorithm and the bisecting K-means clustering algorithm are analyzed,and an improved k-value self-determination and clustering center optimization algorithm is proposed to overcome the shortcomings of the K-means clustering algorithm and the dichotomous K-means clustering algorithm.In order to verify the validity and accuracy of the improved algorithm,UCI data set is selected as the test data,and the clustering results are compared with those of the traditional algorithm,which proves that the improved algorithm has better stability and accuracy than the traditional algorithm.Thirdly,according to the model of power selling packages in China's electricity market,a soft set model for preliminary screening of power selling package is established,and the algorithm for preliminary screening of power selling packages based on soft set theory is determined.Then,the improved bisecting K-means algorithm is used to cluster power selling packages.The results show that this method can solve the multi-objective optimization problem of power selling packages accurately and effectively.Finally,this paper designs and develops an intelligent electricity selling system from the aspects of system demand analysis,system structure design,system function module realization and system testing.Dual soft set theory and improved dichotomous K-means algorithm are applied to the selection module of electricity sales package,and good selection results are obtained.Intelligent power selling system mainly realizes three sub-modules: subscriber side(package selection,customer service,personal center),subscriber side(package management,contract management),system management module(market area management,market member management,package audit,contract supervision,registration,login,password modification).It provides convenience for users to choose electricity sales packages,and improves the market competitiveness of electricity sales companies.