Fault analysis of multiple transmission lines based on density-based logistic regression

The effective and accurate fault analysis of transmission lines play an important role in the stable operation of the power system. In this paper, we propose a novel multi-classification model based on density-based logistic regression (MCDLR) to solve the asymmetric fault analysis of multiple transmission lines. According to the Nadaraya-Watson density estimation, the proposed model maps the training data into a feature space in which an optimization model can optimize the feature weights and the bandwidth of Nadaraya-Watson density estimation. Experimental results show that the accuracy of MCDLR is above 90% or even more.