基于Pearson相关系数与MLP的电费拖欠风险预警算法

Pearson Correlation Coefficient and MLP-based Charge Arrear Risk Alert Algorithm

用户电费回收风险预警是电力公司运营中的一个难题,为此提出了一种基于多层感知机(MLP)的电费回收风险预警方法。首先,利用Pearson相关系数对影响电力客户拖欠电费的因素进行特征提取;然后,提出并描述了基于MLP的电费回收风险预警具体流程;最后,在一组实际数据上与传统的logistic回归算法对比,证明该方法的高效性。对比结果显示,该算法的预测精度达到了92.31%,可为供电单位在风险客户管理上提供预警。

The risk warning of user electricity bill recovery is a difficult problem in the operation of power companies.This work presents a risk warning method for multi-layer perceptron(MLP).The Pearson correlation coefficient is used to extract the characteristics of the factors of electricity charge recovery risk warning based on MLP.In addition,comparison of the proposed algorithm with the conventional logistic regression algorithm on a set of real data indicates its high performance.The results show that the prediction accuracy of the proposed algorithm reaches 92.31%,which can provide early warning for power enterprises in the management of electricity customers.