Fault Analysis Method Of Distribution Network Based On WAMS/SCADA Data

The power distribution network is usually located at the end of the main network,for all power users quality electricity is provided.The reliability of power supply premise is to be ensured by power distribution network fastly and accurately.The economic losses caused by the failure and the impact on the life of the residents can also be reduced.So far,a lot of research results have been obtained for the analysis of distribution network.However,there are still some shortcomings that need to be studied.When there is a fault in the distribution network,a large number of fault measurement information has been poured into the dispatch center.By configuring the PMU,the compatibility is acieved by WAMS system data and the SCADA system,this method helps the estimation accuracy to be improved.Meanwhile,the estimated error can be reduced when the load changes greatly.Compared with the existing methods,the accuracy can be improved.In order to make fault analysis speed and accuracy improved,the distribution network fault analysis based on wavelet entropy is proposed.Wavelet analysis and entropy are combined,their advantages are fully reflected.Wavelet energy spectrum entropy,wavelet fault entropy,wavelet singular entropy are defined to show electrical signal phase angle larger mutation characteristics due transition from normal state to fault state in distribution network,so that fault components are accurately geted.Considering of the fault model adaptability and portability,distribution network failure analysis method is proposed based on directional layered fuzzy Petri net.Fault diagnosis model based on directional protection set form is proposed.Probable fault component failure probability has been obtained through hierarchical transition function.At last,a distribution network fault phase identification method is proposed based on fuzzy Petri net.Characteristic time series of 3 IMF components are dispatched by fault element’s three-phase current and zero sequence voltage through EEMD algorithm.New fuzzy Petri net initial location is entered through the corresponding fuzzy rules quantification,then fault type is obtained according to matrix algorithm.