Smart Transformer Online Monitoring And Fault Diagnosis Technology

With the rapid development of the power demand of the rapid increase of power system as well as the power industry facing improve the utilization rate of power equipment, reduce the running cost of electric power equipment and enhance the serious challenge to the safe and stable operation of the power system. The transformer is one of the most important equipment in the power system, as accurate monitoring and diagnosis of the fault occurs, it is very important to improve the safety and reliability of power system.On the basis of new traits and needs of the in-depth study of the background of smart grid transformer, the difficulties and challenges faced by the current transformer fault diagnosis, to study and sort out the large transformer main failure types and causes, further discuss current based on the analysis of dissolved gas in oil of transformer fault diagnosis of conventional methods, and points out the main problems existed in conventional oil dissolved gas in fault diagnosis method.According to the three ratio method and ordinary neural network model diagnosis of power transformer based on common problems and shortcomings, this paper proposes a genetic algorithm of improved wavelet neural network model based on, will reflect the transformer insulation oil element characteristics and oil temperature, oil pressure, all kinds of noise signal collects a sample set, the neural network was trained, the network is put into actual operation can be for a variety of possible faults make correct judgment. Experiments show that the improved wavelet neural network with the traditional three ratio method, neural network and wavelet neural network method has more advantages compared to the accuracy of fault diagnosis, and can make up for the coding and critical value criterion defect can’t diagnose the faults of the problems exist in the traditional three ratio method.Finally based on the construction of transformer improved wavelet neural network model, completed the design of intelligent transformer online monitoring and diagnosis system is developed for, from condition monitoring, monitoring of partial discharge, gas in oil and micro water online monitoring and online monitoring of winding temperature five aspects are introduced in detail intelligent transformer on-line monitoring scheme, further from respectively from design decisions, input three output and process design point of view followed by the introduction of the specific development method of intelligent transformer online monitoring and fault diagnosis software for each functional unit.In this paper, for intelligent transformer development present situation and the development demand, from the transformer engineering practice of focuses on the background of smart grid transformer online monitoring and fault diagnosis technology, to further address the current transformer oil spectrum analysis diagnosis method, the design and development and supporting the online monitoring and intelligent diagnosis system, effectively improve the transformer in power grid operation security and reliability.