Reliability Prediction of Distribution System Based on Rough Set Theory and Neural Network

With the growing scale of power distribution system, the network frame structure is increasingly complex, and the traditional reliability prediction methods are faced with different degrees of limitations. The application of the big data technology has provided a new way to solve this problem. This paper proposes a reliability prediction method of distribution network based on rough set theory and neural network. First, according to the correlation analysis method based on rough set theory, the correlation index of information entropy was established to extract the strong correlation factors as input to the prediction model. The correlation index of information entropy is established, to extract the strong correlation factors as input to the prediction model. Then the neural network is trained by historical data samples to obtain prediction model, and the genetic algorithm is used to optimize the initial weight and threshold of neural network, in order to improve the convergence speed of neural network and realize a rapid prediction of distribution system reliability. Finally, taking Beijing yearbook data as an example to verify the method, and the results show that the method is effective and operable.