Fault Diagnosis Method Of High Voltage Circuit Breaker And Communication Network In Intelligent Substation

With the rapid development of China's economic level and social life,users and enterprises have continuously increased the requirements for power quality.At the same time,the extensive application of information technology and intelligent algorithms such as big data and cloud computing in various fields has also given rise to the intelligent development of the power grid and further improvement in the level of safe and stable operation of the power system.At present,in order to promote the development of smart grids,China has built more than 1,000 pilots of smart substations.The biggest features of smart substations are the intelligent equipment,the communication network,and the standardization of data information,which makes the various types of information that characterize the operating status of the station more comprehensive and highly shared,and provides conditions for rapid and efficient fault diagnosis and repairment of important functional structures in the station.The important functional structures in smart substations include relay protection devices,secondary control systems,high-voltage circuit breakers,and so on.At present,there are many research results on the reliability of relay protection and communication networks in smart stations,but under the new technical characteristics of smart substations,there is less systematic research on fault diagnosis of high-voltage circuit breakers.The failure of the high-voltage circuit breaker will seriously affect the safe and stable operation of the power system,and the timely and accurate analysis of the failure cause is the basic issue to ensure the safe operation of the power grid and the realization of the self-healing function of the smart grid.Therefore,this article takes the alarm information in the intelligent substation as the analysis object,and proposes a system-complete fault diagnosis method for the high-voltage circuit breaker,which considers the incorrect operation of the circuit breaker caused by the failure of the high voltage circuit breaker itself and the related communication network.The specific work and results obtained are as follows:In view of the lack of systematicness of the existing high-voltage circuit breaker fault diagnosis methods,this paper proposes the overall architecture of high-voltage circuit breaker fault diagnosis based on alarm information,which is combined with the new technical characteristics of intelligent substations and the actual fault information of high-voltage circuit breakers.The structure includes the analysis of abnormal information in the station,the method for judging the fault condition of the circuit breaker preliminarily,and the fault diagnosis methods for the high-voltage circuit breaker itself and the related communication network.Among them,the anomaly information analysis method proposed in this paper uses the BMH(Boyer-Moore-Horspool)algorithm and the semantic frame mode to realize the screening of target information and the acquisition of important information.Based on the VSM,the information in the semantic frame is vectorized to make it directly applied to fault diagnosis algorithms and achieve automatic connection of data and algorithms.The case analysis in the fault diagnosis algorithm proves the feasibility and practicability of the proposed architecture and method.Because the existing fault diagnosis methods of high-voltage circuit breakers are mainly based on the analysis of micro-quantities such as the opening and closing coil currents and mechanical vibration signals,without considering their own physical relationships,diagnostic coverage and accuracy are limited.Based on the analysis of the operation process of the high-voltage circuit breaker and the mapping relationship between the alarm information and the fault reason,this paper proposes a circuit breaker fault diagnosis method based on a weighted fuzzy Petri net model.The proposed method combines Petri nets with fuzzy theory and introduces concepts such as virtual place and negative propositions,to realize the correlation of the five operating phases of the high-voltage circuit breaker,which incorporates the physical relationships inside the circuit breaker into the diagnostic model.Through actual case analysis and verification,the method can effectively and comprehensively diagnose high-voltage circuit breaker faults,and obtain diagnostic results covering various possible fault causes.The existing research mainly focuses on the fault diagnosis of the high-voltage circuit breaker body,and lacks of research on incorrect operation of circuit breakers caused by failure of other parts when the body is normal.The incompleteness of the fault cause analysis will not only cause maintenance work to become a bottleneck,but also makes it difficult to fully advance the self-healing function of smart grids.Focusing on the situation that the high voltage circuit breakers incorrectly operate due to related communication network faults,this paper proposes a fault diagnosis method based on the directed bipartite graph model,so as to trace the specific fault causes back.According to the transmission path of the GOOSE message,the related communication systems are clustered into communication networks,protection devices,intelligent components,and measurement and control devices.Each cluster is regarded as a subdomain,in which a directed bipartite graph model is built based on the logical relationship between its alarm information and the cause of the failure.By introducing the Bayesian likelihood,the cause of the failure is calculated and analyzed,and then it's verified with reverse reasoning.Therefore,the accuracy of the result is guaranteed from the probability.Case analysis shows that the fault diagnosis method can trace the source of the incorrect action of the high-voltage circuit breaker in time,and has high fault tolerance.