Intelligent Community Research and application of online monitoring and fault diagnosis method for distribution Transformers

Intelligent Community Research and application of online monitoring and fault diagnosis method for distribution Transformers

Northeast University Lou Xue Ning

The main work of the thesis is as follows: (1) from the existing problems of the distribution equipment in the community, the current transformer fault diagnosis methods are summarized. (2) According to the complex non-linear relationship between the equipment running state and the fault phenomenon, the BP neural network has very good complex non-linear mapping function, so the BP neural network is used to diagnose the distribution transformer. (3) According to the requirement of transformer fault diagnosis, BP network structure is designed. In order to solve the problem that BP neural network is slow in convergence and easy to fall into local minima, genetic algorithm is combined with neural network to compose genetic BP network hybrid algorithm. The algorithm utilizes the global search ability of genetic algorithm, optimizes the connection weights and thresholds of neural network, makes the network get good initial value, improves the disadvantage of BP network and improves the accuracy of fault diagnosis. (4) According to the functional requirements of the Intelligent Community Service integration System for the monitoring and control subsystem of the distribution substation equipment in the community, the subsystem of on-line monitoring and fault diagnosis of the distributing transformer is designed and implemented, and the application of BP network in the fault early warning system is realized.

Intelligent Community Research and application of online monitoring and fault diagnosis method for distribution Transformers