电力客户信用动态智能分析系统的设计及实现

摘要

In order to evaluate the credit level of power customers and enhance the ability of controlling risk,,a credit quan-titative model depending on the analytic hierarchy process (AHP) is proposed in this paper. Based on this model, the Logistic regression model is used to calculate the probability of performance and the probability of default. The loss of default is calculated through the probability of default and the power consumption. The risk early warning model based on the probability of default, the loss of default and the decreasing degree of the credit scores was built. In consequence, the risk dynamic intelligent analysis system for power customer credit score was created. A perfect social benefit and economic benefit are gained with this system.%为能更加有效地评估电力客户的信用水平,增强供电企业事先风险控制的能力,在此通过层次分析法,建立了信用量化模型.在此基础上,运用Logistic回归模型计算用电客户的履约概率和违约概率.然后通过综合考虑违约概率和用电量,计算违约损失.最后从违约概率,违约损失以及信用下降程度3个方面进行风险预警,建立了电力客户信用风险动态智能分析模型,进而建立了基于此模型的电力客户信用智能分析系统.该系统在业务领域创造了很好的社会效益和经济效益.