The Technology On The On-line Monitoring And Fault Diagnosis Strategy Of Intelligent Circuit Mechanical Properties

High voltage circuit breaker is the most important protection and control equipment of substation, who’s heaith status is closely related to the safe and stable operation of power system. In order to guarantee the reliable operation of power system,the strict requirements of circuit breaker operation quality are put forward.Therefore, the intelligent circuit breaker on-line monitoring and fault diagnosis technology arises at the historic moment.The technology can grasp the operate status of high voltage circuit breaker real-time and make the fault diagnosis aim at the appeared abnormal situation so as to the timely maintenance,which is playing an important role operation reliability of the power system.First of all,This paper introduces the methods of calculation and analysis of mechanical characteristic parameters are put forward which is aim at the principle of monitoring.This paper aimed at the TMS320F28335 chip proposed the design of software and hardware aim to the intelligent circuit breaker on-line monitoring device,This chip integrated with many function modules internally,which is reduce the design of PCB.Intelligent circuit breaker on-line monitoring device integrated software program that can collect reliable mechanical parameters of the monitoring quantity which were have mentioned above and simply introduced the original data preprocessing method of acquisition parameters reliable.The device of circuit breaker mechanical properties on-line monitoring can collect circuit breaker energy-storage electric current, the breaking and closing coil current, dynamic contact travel-time characteristic, the auxiliary contact status data mechanical parameters and so on.Secondly, aim at the complexity and uncertainty of intelligent circuit breaker operation, this paper proposes a kind of fault diagnosis algorithm of RBF neural network and rough sets.This part first introduced the synthesis principle of rough set and RBF neural network, then the reduction of intelligent circuit breaker diagnosis decision table using the method of attribute reduction in rough sets theory, Finally using the RBF neural network which is established by minimal diagnosis decision table to make comprehensive diagnosis of intelligent circuit breaker, to ensure the accuracy of intelligent circuit breaker fault diagnosis.In the synthesis algorithms, rough set can simplify the complex networks made by the amount of mechanical characteristic parameters of on-line monitoring equipment, and can satisfy the actual needs of the minimalist network, thereby reducing the training time, and to a great extent, improved the diagnosis ability of the network.The device of circuit breaker mechanical properties on-line monitoring has been successfully installed in the substation of Goushan, Jiangshan,Zhejiang, and can be stable operation.The rough set and RBF neural network algorithm which is fusioned by fault diagnosis expert system software in intelligent substation in Hongqiao could make a accurate diagnosis to the health status of the circuit breaker.Therefore,the on-line monitoring and fault diagnosis technology of intelligent circuit breaker can be applied in practical engineering well and have a great significance to the safety and reliable operation of power system.