| With the development of science and technology,cloud computing which has the high reliability and large data processing capability has become a research hotspot of IT and also achieved great success as a new computing model.However,due to the development of smart grids and distribution network,the existing computing platform has been unable to meet the needs of the large-scale computing,and even has become the bottleneck of the development of smart grids.Because of the seriously contradiction,new technologies are required to solve this problem.In view of the high efficient data processing and storage capacity of cloud computing,an interdisciplinary approach is proposed to solve the problem of distribution network reconfiguration using the cloud computing technology in this work.At present,cloud computing is rarely utilized in electricity industry,so this research will have a great effect on the development of the smart grid.The research status of the distribution network reconfiguration and cloud computing technology in the power system has been discussed in detail firstly.Then the concept,characteristics and the development of cloud computing have been recommended,the working principles and processes of the open source software Hadoop have been analysed.In the meantime,Particle Swarm Optimization(PSO)has been used to solve the mathematical model of the distribution network reconfiguration problem which we built in this study.In the solving process,we have proposed a new connectivity search algorithm and used a new fast power flow solution algorithm.This new algorithm have accelerated the speed of the flow calculation and improved the efficiency of the algorithm,the efficiency has been tested by the example of the simulation.According to the parallel features of the PSO,the cloud solution of the PSO has been designed and realized with Java language on the Hadoop platform.The experiment results have demonstrated that the cloud solution of PSO is a efficient measure to optimize the distribution network reconfiguration problem.The methods were improved according to the disadvantages coming out in the experiment process. |