An Optimization Scheme Of Tiered Electricity Pricing For Regional Residents Based On Improved Clustering Algorithm

Power enterprise in China implements an incremental tiered electricity pricing for residential consumers.According to the accumulated electricity consumptions in the billing cycle,the residential tiered electricity pricing system measures the electricity bills differently,and guides consumers to participate in energy conservation and emission reduction and reasonably plan the electricity consumption behavior.However,the current scheme has a long renewal period,and the existing research and formulation methods also have their own defects.It is necessary to design a new optimization scheme according to the actual situation of the region and local conditions.The main work of formulating the scheme of residential step price is to design the parameters of tiered level,tiered length and tiered unit price.The promotion of smart grid and the development of big data technology lead the power industry into the era of power big data.By mining the hidden information in the power data to solve the practical problems,it has changed the previous thinking mode based on the conditions of intuitive or shallow hidden information,making it possible to deal with the problems with complex data and unknown model.By applying the appropriate big data analysis method to the field of power economy,it provides a new idea for the formulation of residential tiered electricity pricing optimization scheme.Firstly,the concepts of data mining,clustering analysis and swarm intelligence algorithm are introduced.According to the characteristics of power consumption data and the target task of this paper,the k-means algorithm is chosen as the main data analysis method.Considering the shortcomings of the algorithm in the selection of cluster number and initial cluster center,this paper improves the algorithm based on the advantages of other algorithms.The improved clustering algorithm overcomes the adverse effect of random parameters,improves the accuracy and reliability of clustering results,and provides more reliable support for guiding to solve practical problems.Secondly,two improved clustering algorithms are tested based on UCI dataset.Four representative datasets,such as the number of samples and data dimensions,are selected as the experimental objects for clustering processing,and the cross combination experiments of different types of datasets and different improved clustering algorithms are carried out.By analyzing the corresponding clustering results,the optimal use scenarios of the algorithm are summarized to better deal with the actual clustering problems.Finally,an optimization scheme of residential tiered electricity pricing is designed based on the actual electricity consumption data.The k-means algorithm based on the adaptive weight PSO is used to cluster the collected annual electricity consumption data of residents.Based on the optimal clustering results,the optimization scheme of tiered electricity pricing is formulated.Compared with the current scheme,the optimization adjustment scheme can better play the role of regulation and guidance of residential tiered electricity pricing,and the optimization strategy is simple and effective,which has a certain reference value for the optimization research of residential tiered electricity pricing...