Study On The Power Flow In Power Systems Based On Cloud Computing

With the rapid development of China’s social economy,demand and supply of power are growing rapidly.the robust of power infrastructure construction has kept increasing with complexer structure and larger scale of the more connected power grid.This lead to a bigger demand in computing and storage of power systems,so the study in application of cloud computing in power systems and building a power system based on intelligent computing platform is an inevitable choice.As the most basic and most important calculation problem in power system analysis,power flow calculation is also faced with the problem that it is difficult to calculate accurately and quickly.Therefore,this paper studies the cloud computing method of power system power flow.Firstly,this paper studied the main features of cloud computing.it also analysed the read and write flow of Hadoop HDFS and the data processing of MapReduce.Then the characteristics of Spark RDD and the operating mechanism of Spark are summarized,which provides a theoretical basis for the follow-up study of the paper.Secondly,this paper points out that the essence of Newton-Raphson method of power flow calculation is the transformation from nonlinear problem in solving the power flow equation to the linear equation.The difficulty lies in the solution of the linear modified equations.By comparing the Cimmino method,the ART method and the SART method,a power flow calculation method based on the Newton Raphson method and the ART method(Newton-Raphson-ART Method)is proposed.This new method using ART method to conduct the iterative solution of the linear modified equations.it can be easily used in parallel calculation of the power flow in the cloud envrionment.All the simulation result shows that the proposed method can effectively solve the power flow problem and the result is accurate.At last,base on the proposed method,this paper designed an algorithm to calculate power flow problem with building Hadoop and Spark cluster environment,and tested it in many different IEEE standard test systems.Result shows that this algorithm can be parallelly executed and has a good speed in power flow calculation.