Research On Transformer Fault Diagnosis Based On The Cyber-Physical Systems

Power transformer in the power transmission and distribution system to undertake the conversion voltage level,the important function of power transmission is an important support for the safe operation of power systems equipment.Transformer in the operation of any accident will bring economic losses,and even lead to serious social impact.At present,the artificial intelligence algorithm is mostly used for the fault diagnosis of transformers.Although the introduction of artificial intelligence algorithm greatly improves the deficiencies of traditional diagnosis methods,the accuracy of fault diagnosis has been greatly improved,but many of the fusion algorithms Scholars simply use the merits of various algorithms to fuse,resulting in redundancy and mutual exclusion of the system.In terms of troubleshooting,man-made routine inspection has certain delay problems in finding faults and handling faults.Therefore,researching fast and accurate transformers Fault diagnosis techniques and methods,and the timely and effective elimination of faults,the safe operation of the grid is of great significance.In order to solve the existing problems of transformer fault diagnosis,a fault diagnosis model based on support vector machine is proposed firstly,and a particle swarm optimization algorithm is introduced,which is deeply optimized by random weights.The SVM algorithm solves the problem of transformer fault Type and fault features,and it has an absolute advantage in dealing with small sample data.However,the support vector is more sensitive to the choice of parameters and lacks the standard basis for choosing parameters in practical applications.Therefore,The particle swarm optimization algorithm with good adjustability is optimized and random weights are introduced to effectively improve the premature and local optimal particle swarms.The deep optimization algorithm has greatly improved the accuracy of the transformer fault diagnosis model.Secondly,design a transformer information physics fusion system based on deep optimization model as the processing center.Cyber Physical Systems is a complex system of integrated computing,communication network and physical entities with real-time sensing,dynamic control and information service Features.Applying this system to transformer fault detection,it can timely analyze the operation characteristics of the transformer and effectively solve the problem of delay in operation during the fault diagnosis.The fault diagnosis process of transformer based on the Cyber-Physical Systems mainly collects the transformer status data through the physical equipment installed in the transformer and transmits the data to the grid control center through the transmission network.The control center adopts the deep optimization fault diagnosis model to further improve the data Processing,draw the judgment result and make the decision of the corresponding fault,realizes the real-time monitoring to the running status of the transformer.The simulation results show that the deep optimization model has higher diagnostic accuracy.