Research On Fault Location Method Of Active Distribution Network Based On Wide Area Information

In recent years,with the continuous growth of the national economy and the access of distributed power sources such as photovoltaic and wind power,active distribution networks have been more widely used in China.Due to the access of large-scale distributed power generation,the flow of power flow in the distribution network has changed from one-way flow to two-way flow.At the same time,when the active distribution network fails,the complexity of fault current,weak fault characteristics and other reasons increase the difficulty of fault diagnosis and location,which may lead to the failure of the original fault location technology to be directly applied to the active distribution network.Therefore,the study of fault location methods applicable to active distribution network has become a major issue that must be tackled in practical engineering projects.(1)The fault location method of active distribution network based on the over-current distribution law is realized.For the active distribution network with multiple measuring points,the over-current direction on each feeder is obtained through data acquisition and processing.Based on the distribution law of fault overcurrent and the shortest path theory in graph theory,the suspected fault area is determined to reduce the size of the solution space used for fault location,and a new fitness function is proposed for fault location to make it more sensitive to distortion information.In the solution space,the improved binary marine predators algorithm(IBMPA)is used to search the fault feeder,and the ability to jump out of the local optimum and the speed of global search are improved through chaos mapping,adaptive t distribution and group learning strategy,so as to achieve accurate fault location.(2)An active distribution network grounding fault location method based on zero-sequence current distribution law is realized.Single-phase ground fault in active distribution network is characterized by weak and complex features,which is more likely to cause distortion of fault information collected at measuring points.Fault location based on local or local features is easily affected by information distortion,resulting in difficulty in location.Referring to the application of intelligent algorithm in the location of inter-phase short circuit fault in distribution network,a method of active distribution network ground fault location based on the wide-area distribution of transient zero-sequence current is proposed,which is less affected by information distortion and accurate.(3)The fault location method of active distribution network based on multi-core graph attention network is realized.There have been a lot of research results in the power grid fault diagnosis technology based on artificial intelligence,but the distribution network topology changes frequently,and the traditional artificial intelligence methods highly rely on training data,which brings difficulties to the fault location of distribution network.To solve the above problems,a distribution network fault location method based on multi-core graph attention network is proposed.The electrical nodes and lines of the distribution network are mapped to the vertices and edges of the graph in the graph attention network.The attention coefficient is calculated according to the similarity of the fault features between adjacent vertices,and the correlation between the vertex features is better integrated into the fault location model,which improves the adaptability of the fault location model to topology changes.Finally,a distribution network fault simulation model is built to verify that the proposed method has the advantages of high positioning accuracy,good robustness and is not affected by fault resistance,fault initial phase angle and fault distance,and the good application effect of the model in actual comprehensive fault scenarios is verified under different network topology changes and scenarios.