| With the fast extension of power systems and the trend of buildingsmart grid which is defined as "information, automation, interaction", thetraditional power system information platform is inadequate in theintegration, analysis, storage and computation of information, and newtechnologies are required to solve those problems. Cloud computing is anew scientific concept that developed from technologies such as distributedcomputing, virtualization and network storage. It has characteristics ofhigh performance and lower cost, can effectively meet the needs of futuredevolopment of power grid, and can be used to build the new intelligenceplatform of power systems.As so far, there are few literatures on the cloud computing in powersystem application, and the power system cloud computing platform isnothing but a plan. Because power flow calculation is the basis of stabilityanalysis, and play an importment rule in the advanced application ofinformation platform, cloud computing based power flow algorithm ismeaningful to the success application of cloud computing platform inpower sysytem.At first, this paper introduced the concept and characteristics of cloudcomputing, and studies the background and key technology of extensiveopen source software Hadoop. Second, the parallel algorithms of powerflow based on CAV and Jacboi iterative methods were designedrespectively by. And then, theHadoop experimental platform is constructedand at last, the effectiveness and feasibility of the algorithms were verifiedand discussed by examples on the platform. |