Research on the Electricity Customer Credit Evaluation Based on Fuzzy Expected Value Decision-making Method Modified by Least Squares Support Vector Machine

Abstract

In view of electricity customer credit evaluation lacking of precise index system and hardly quantifying subjective factors and experience factors, fuzzy expected value decision-making method modified by least squares support vector machine (LS-SVM) is presented. Firstly, electricity customer credit evaluation index system is constructed; the indices values and subjective experiences values are given in the form of triangular fuzzy numbers. Then credit expected values are resulted by fuzzy expected value decision-making method. Finally, LS-SVM based on the principle of structural risk minimization modifies the expected values. The experiment shows that the credit grades after the modification suit to the original credit grades enacted by power supply enterprises and are more practicable.