论文部分内容阅读
电费收缴工作的好坏,将直接影响电力公司的经营状况。为了约束电力客户的用电行为,保证电费的及时收缴,事先准确地识别客户的信用状况,是最为关键的内容。考虑到电力公司对客户信息的掌握情况以及信用管理的要求,设计了一套符合实际管理需要的,能够反映电力客户实际信用水平的指标体系,并在此指标体系的基础上利用自适应神经模糊推理系统(ANFIS)建立信用评级模型。该模型经测试样本检验证明是精确可靠的,适用于实际配电管理中对电力客户信用类别的判别。从而为电力公司预知信用风险,并采取不同的信用管理对策,提供了有力的参考依据。
Electricity fee collection work will be a direct impact on the power company’s operating conditions. In order to restrain the customer’s power consumption behavior, to ensure timely payment of electricity to accurately identify in advance the customer’s credit status, is the most crucial content. Considering the power company’s grasp of customer information and credit management requirements, this paper designs a set of index system which can meet the actual management needs and can reflect the actual credit level of electric power customers. Based on this index system, an adaptive neuro-fuzzy Inference System (ANFIS) Establishes a credit rating model. The model tested by the test sample proves to be accurate and reliable, and is suitable for discriminating the power customer credit category in actual distribution management. So as to predict the credit risk for power companies, and take different credit management strategies, provide a strong reference.