Research And Application On The On-line Monitoring And Fault Diagnosis Method Of The Distribution Transformer In Intelligent Community

The rapid development of the internet of things technology makes this technology widely used in Intelligent Transportation, Public Security and other areas. Intelligent Community is a new type of community, which use the Internet of Things, Cloud Computing and other new technologies to build a safe, convenience, modern and intelligent community. Distribution transformer is the final link of power supply to the residents, the stability of power distribution system is related to its operation state, and the transformer plays an important role in the power distribution system in the community. Therefore, it has a great significance to monitoring the transformer effectively and warning faults.Based on the study of the distribution transformer monitoring and the development of fault diagnosis technology, and combined with the function demand of the Intelligent Community project, proposed a method that using Genetic Algorithm converged BP neural network hybrid algorithm for fault diagnosis, and use Internet of Things Technology to monitoring the operating states of the equipment at the same time. So, we can find the abnormal operation information of distribution transformer timely and to control or resolve the problem, this ensures the power system in a security and stability operating environment. The following is main work of the paper:(1) According to the practical problems of power distribution equipment in community, introducing some transformer fault diagnosis methods.(2) Think over the nonlinear relationship between the operation state and the fault types, and BP neural network is good at nonlinear mapping, so, we diagnose faults by using BP neural network.(3) In connection with the demand of transformer fault diagnosis, designed the structure of BP network. In order to solve the shortcomings such as slow training speed and inclined to immerse into the partial least point of the BP network, there is another way that combined genetic algorithm with BP network which is called hybrid algorithm. By using the global searching ability of genetic algorithm to optimize the weight and threshold, so the network can have a better initial value, to improve the shortage of BP network and the diagnostic accuracy.(4) Based on the function demand of intelligent community service integration system and the power distribution equipment supervision subsystem, designed and implemented the distribution transformer on-line monitoring and fault diagnosis subsystem, and also implemented the application of BP network in the fault diagnosis subsystem.