Research Of Power Grid Fault Diagnosis Based On Multi-Source Information Fusion

With the constant expansion of the size of power grid,the connection between different regions is getting closer.When a power grid fails,the impact of faults on the system will be increased.Existing power grid fault diagnosis methods are mostly based on breaker tripping,protective action and other switch value information.However,such a method as processing switch value using a given intelligent algorithm to get fault diagnosis results has a high requirement for the integrity and accuracy of breaker and protection information.Different degrees of integrity and accuracy have a great effect on diagnosis results.This paper explores a power grid fault diagnosis method using multi-source information.First of all,the research background and current situation at home and abroad is introduced.Secondly,all kinds of fault information sources in current grid fault diagnosis are introduced.Next,grid fault diagnosis based on switch valve information is analyzed.In view of network adaptability and uncertain information fault tolerance of the existing Petri net diagnosis model,a sequential weighted fuzzy colored Petri net diagnosis model is proposed.Meanwhile,this paper analyzes grid fault diagnosis based on electric quantity.In view of over-enveloping and under-envelop in fault current analysis of HHT,an improved empirical mode decomposition mode is proposed.The improved HHT process can generate more accurate diagnosis results.On the basis of the above method,this paper puts forward a grid fault diagnosis fusion model based on D-S evidence theory and a diagnosis and decision model based on C-means algorithm.Finally,based on Matlab software,through a simulation analysis of comprehensive cases,the effectiveness of the proposed method is verified.