Electric power user credit evaluation method based on global optimal fuzzy kernel clustering model

Abstract

The invention discloses a power user credit evaluation method based on a global optimal fuzzy kernel clustering model, comprising: establishing a power user credit evaluation index system, and performing preprocessing on evaluation index data; establishing a global optimal fuzzy kernel clustering algorithm model; Input the preprocessed power user credit evaluation data to be evaluated into the established global optimal fuzzy kernel clustering algorithm model to obtain the clustering results; perform qualitative analysis on the obtained clustering results to obtain the credit rating of each category; extract The characteristic index and corresponding characteristic value of each cluster in the clustering result; according to the extracted characteristic value index and corresponding characteristic value of each cluster, evaluate the credit result of the power user to be evaluated again. It replaces the manual evaluation of the power user's credit, reduces the input of manpower, saves the management cost of the power enterprise, and improves the scientificity and accuracy of the evaluation work.

Description

technical field [0001] The present disclosure relates to the technical field of information processing, in particular to a power user credit evaluation method based on a globally optimal fuzzy kernel clustering model. Background technique [0002] With the rapid development of the country's economy and energy industry, power users' demand for electric energy is increasing. In the process of power marketing for power supply companies, the credit status of power users will have a direct impact on the operation of power companies. In the current huge power user group, there are quite a few users who are weak in anti-risk ability, prone to poor capital turnover, unable to pay bills on time, resulting in the occurrence of similar credit loss phenomena such as power theft and arrears. These phenomena have brought power enterprises Therefore, how to evaluate and grasp the credit situation of power users in a timely manner and effectively avoid operating risks is a practical problem...

AI Technical Summary

Problems solved by technology

In the current huge power user group, there are quite a few users who are weak in anti-risk ability, prone to poor capital turnover, unable to pay bills on time, resulting in the occurrence of similar credit loss phenomena such as power theft and arrears. These phenomena have brought power enterprises Therefore, how to evaluate and grasp the credit status of power users in a timely manner and effectively avoid operating risks is a practical problem that power companies need to solve urgently.

[0003] After years of development, the credit evaluation work has developed from the initial manual evaluation to the information-based calculation evaluation. The inventors found that most of the information-based evaluation processes in recent years are based on subjective scoring methods and subjective and objective index weighting methods. The efficiency of large sample data is low, the evaluation process is also lack of scientific nature, and the algorithm complexity is high; in addition, in the past credit evaluation process, the index characteristics analysis of the evaluation object is often not accurate enough, the lack of accurate processing of this index and the characteristics The lack of analysis can easily lead to relatively large errors in the evaluation results

Images