Research On Fuzzy Petri Net Fault Diagnosis Method Considering Time Information For Power System

With the increasing scale of power grid,when a short-circuit fault occurs on a certain line of the power grid,if it is not diagnosed and restored quickly,the fault area will spread,which will cause a large-scale blackout and cause serious economic losses.Therefore,it is necessary to study an efficient fault diagnosis method for power system to diagnose the fault components in time and send to dispatchers to provide auxiliary decision-making.The fault diagnosis method based on Petri net has the advantages of simple reasoning,clear physical meaning and fast diagnosis speed.Besides,it can represent the logical relationship among power system components,protection and circuit breaker in the form of reasoning diagram,so it is widely used in the field of power system fault diagnosis.On this basis,scholars further proposed fuzzy Petri net to solve the incompleteness and uncertainty of alarm information.Firstly,the research background and significance of power grid fault diagnosis methods,as well as the research status of various fault diagnosis methods are described.The basic Petri net and fuzzy Petri net are introduced in detail,and the connection analysis method used to determine the suspicious fault component is studied,so as to provide a basic basis for the establishment of the later diagnosis model.Secondly,a hierarchical fuzzy Petri net fault diagnosis method considering time constraints is proposed.The principle of power grid protection is analyzed.Based on this,a hierarchical fuzzy Petri net model of suspicious fault components is established,which can adapt to the change of network topology.The protection and circuit breaker alarm information is checked by reverse and forward time sequence reasoning,and the confidence probability of the place which does not meet the time constraint is corrected.The reasoning process and matrix reasoning algorithm of power system fault diagnosis method based on hierarchical fuzzy Petri net are studied.Gaussian function is introduced to modify the probability in the matrix reasoning process,so that the fault probability is always maintained at 0-1,and finally the confidence probability of the fault component and its time point constraint are obtained.The correctness and rationality of the proposed method are verified through the analysis and comparison of power grid system examples.Finally,since the existing fault diagnosis methods based on Petri net do not make full use of the time sequence attributes of alarm information,and the time sequence reasoning and fuzzy reasoning algorithm are complicated,an improved hierarchical fuzzy Petri net fault diagnosis method for power grid considering time sequence is proposed.The improved hierarchical fuzzy Petri net model of the component is established.The confidence probability of fault component can be obtained by hierarchical reasoning combined with matrix reasoning algorithm,and the action behavior of the protection equipment is evaluated.The algorithm for simultaneous execution of fuzzy reasoning and temporal reasoning is studied,and the fault diagnosis framework of the proposed method is constructed.The test results of the local grid and IEEE-14 system show that the proposed method can diagnose the fault components of power grid effectively,and has a good application prospect.