Identifying Correlation between Planning Strategy and Reliability of Distribution Networks Based on Machine Learning

Considering the complexity of the influence of the reconstruction measures for distribution network on the precision investment decision-making scheme, this paper adopts a method of data mining to establish the correlation model between the reconstruction measures and reliability evaluation indexes. In this paper, based on the sample data, the correlation mining methods of the reconstruction measures and performance indexes, including multiple linear, back propagation neural networks(BP) and recurrent neural network (RNN) methods, are applied to analyze the relationship between different types of reconstruction measures and performance indexes. Experimental result shows that the BP algorithm is more effective than the correlation model established by multiple linear regression and RNN. Meanwhile, with the increase of sample data, the accuracy of the correlation model established by BP always fluctuates within a small range. Finally, it is found that the selection of different activation functions will also have a greater impact on the accuracy of the BP correlation model.