Design Of Transformer On-line Monitoring And Fault Diagnosis System For Intelligent Substation

With the development of intelligent transformer substation of a higher level of intelligence,substation transformer on-line monitoring and fault diagnosis technology is gradually improved and enhanced.Improving the intelligent level of the transformer can not only improve the efficiency of the grid staff,but also make great influence on the stable opearation of the power system.The problems in the process of transformer opearation can be detected in time by real-time online monitoring on the operation of intelligent transformer substation.Online fault diagnosis system is able to predict the potential transformer faults to promptly identify problems before transformer failure,avoiding unstable operation or even endanger caused by power outages due to transformer fault of the power system.Optimization of genetic algorithm based on BP neural network algorithm was chosed as the method to complete the the transformer design of on-line monitoring and fault diagnosis system by completing corresponding design on hardware and software.Through the study of artificial intelligence diagnostic methods,the BP neural network algorithm was proposed as a means of transformer fault diagnosis,and the optimization of genetic algorithm based on BP neural network algorithm was presented to optimize the design.By operating a simulation on the Matlab software platform for algorithm,results show that the accuracy of the optimized neural network algorithm has been improved.ARM chip LPC2214 CPU was chosed as the core to design the hardware of the data acquisition and transmission system of the transformer oil gas concentration,by transplanting uC/OS-II real-time operating system,and combined with embedded software system the overall function can be achieved.Based on LabVIEW virtual instrument software,PC online monitoring system was fininshed through graphical programming,which completed data acquisition and DGA analysis module,fault diagnosis module which based on neural network and database module,and accomplished the goal of transformer on-line monitoring and fault diagnosis system of human-computer interaction interface.This article on transformer on-line monitoring and fault diagnosis system not only completed the verification the theory of BP neural network algorithm improved by genetic algorithm,but also analysed the actual demand of the system,through embedded hardware and software system,functions of the concrete data collection,information transmission,on-line monitoring,fault diagnosis,database storage,etc were implemented,accomplishing the design of the transformer on-line monitoring and fault diagnosis system,providing concrete solutions for the practical application of artificial intelligent transformer on-line monitoring and diagnosis.