Research On The Fault Diagnosis Method And Repair Strategy Of Distribution Network Based On Cloud Computing Platform

As the last 10 km of power transmission,Power distribution network connected directly to the user,distribution network fault directly affect the reliability of power supply system.Traditional distribution network automation system affected by the operation mode and the network topology,because of the faulty regional isolation quickly enough,the fault location accuracy is low,the fault area power supply recovery problems is not in time.Study fault diagnosis technology fast and exact,and how to make scientific and rapid repair strategy,to reduce the economic loss and construction high reliability of intelligent distribution network has important significance in theory research and engineering practice value.With the data volume trend,in view of the current distribution network fault diagnosis and fault repair grows exponentially,based on the design of the fault diagnosis and fault repair in the distribution network of cloud computing platform architecture and the establishment of distribution network,multi-source information fusion and cloud computing integration information platform,puts forward and repair the fault diagnosis of distribution network hierarchical multi-source information fusion method based on cloud computing platform.First of all,according to the information of distribution network fault prone to themisinformation phenomenon,proposed a multi-source information fusion fault diagnosis method based on.Firstly,weighted average data fusion of multi-source information in data layer.Then using the wavelet transform method of transient voltage and current information extraction,get the relative voltage distortion degree and the relative frequency of the current distribution difference;the voltage phase information for pattern matching method for WAMS system based on pattern matching get suspicious fault trust;using the improved Bayesian networks for the credibility of the assessment of switch information based on the relative degree of fault.Then,the four kinds of evidences are processed by decision fusion after normalization,so as how to realize distribution network fault diagnosis.Finally the example proves that the method can well solve the problem of fault information themisinformation.Second,if fault occurs,all kinds of information system will be uploaded to the control center a large number of fault information,and accompanied by a greater degree of uncertainty and the present situation of the redundancy.The distribution network layered fault diagnosis method based on cloud computing platform is put forward.Firstly switch protection straton cloud to rapid diagnosis of simple failure under the coordination control of total cloud platform,given the amount of diagnostic information source for all kinds of simple switch protection layer,Using Petri net fault diagnosis in a cloud computing platform;When a fault is complex,Using the substation straton cloud cannot complete diagnosis,considering the layer diagnosis multi-type information source,contains all kinds of electric parameters and quantity of the switch protection,Using intuition uncertain fault diagnosis rough set theory in a cloud computing platform.Between two child cloud can be realized dynamic jump under the coordination control of total cloud platform,according to the characteristics of longitudinal diagnosis entry and structure adjustment.And this method can very good deal with large data,the efficiency and accuracy of fault diagnosis are improved,and it has good application value.Last,in the summary of domestic and foreign scholars on economic losses、the path and the repair time optimization,considering the level of fault emergency workers difference and the diversity of the demand for different fault repair,in this paper，considering the repair compatibility problems,puts forward the model of points fault repair methods based on cloud computing platform.In support of the strong data which in the distribution network integration of multi-source information fusion cloud computing information platform，according to upload module diagnosis and fault repair team to repair information first model choice.Then use of cloud computing fast obtain the repair ability,objective conditions and fault repair demand indicators.By comparing calculation repair team ability and the compatibility between fault requirement,to integrate the corresponding fault repair plan formulation,and verify the practicability of method.To ensure the scientific and rapid fault repair scheme,realizes the distribution network fault repair the lean management,improves the inspection department management level.