Research On Fault Analysis Method Of Smart Distribution Grid Based On Improved Petri Net

Smart distribution grid is the inevitable form of distribution network,with high quality,safety,reliability and efficiency.However,its size and complexity will greatly increase the difficulty and vulnerability of operation and control,which brings great challenges to the fault diagnosis of distribution network.When multiple faults occur in the distribution network,the alarm information will increase rapidly,increasing the burden of workers to deal with information and the difficulty of fault diagnosis.Correct and effective fault diagnosis is of great significance for isolating fault components,reducing power outage area,shortening accident treatment time and realizing rapid recovery of distribution network faults.In this article,the basic theory of Petri net and Bayesian network is briefly introduced,and various improved models of Petri net are deeply studied.On this basis,an improved Bayesian Petri net fault diagnosis model considering quantitative temporal association rules(TABPN)is proposed.Firstly,the temporal association rules(TAR)are introduced,the forward and reverse reasoning process of the TAR is described,and the mathematical definition of the TAR is given.The TAR table based on the TABPN is established.The time relationship can be searched and reasoned correctly and effectively based on the TAR table.From a quantitative point of view,the TAR describes the timing relationship and logic rules between fault element,protection action and tripping of corresponding circuit breaker.Secondly,based on the timing characteristics of fault alarm information,the identification method of alarm error information is proposed,and the identification algorithm of alarm information breakpoint,loss and false alarm is given.Whe n the action and temporal information of protections and breakers are received,the identification algorithm is used to identify the mal-function,abnormal information and time scale error.Finally,according to the Bayesian network based on the state estimation method to deal with the information set of incomplete,under the condition of incomplete fault information,it uses the protection action rules to reason the timing constraint relationship of alarm information.The fault probability of each failure chain obtained by Bayesian backward reasoning and the average probability are calculated based on the TABPN,forward reasoning is to obtain the trigger probability of each protection and circuit breaker in the event of the component failure,and the final reduced model is constructed.The example shows that this method is not only accurate and reliable,but also solves the problem of state combination explosion,and further improves fault tolerance of fault diagnosis.