Credit evaluation of electricity sales companies based on an enhanced DHNN model

To standardize the market conduct of electricity sales companies and elevate the power market management level， it is essential to conduct a credit evaluation of these companies. Therefore， based on Box-plot and orthogonalization methods， a credit evaluation model based on an enhanced discrete Hopfield neural network （DHNN） is proposed. Firstly， factors influencing the credit levels of electricity sales companies are analyzed， and a credit evaluation index system， which includes 11 indicators such as basic information， foundational management， contract management， and transaction management， is established. The weights for these indicators are determined using the Delphi method. Secondly， outliers in the indicators of electricity sales companies are addressed to derive optimal credit scores， enabling an objective assessment of their credit levels. Finally， the feasibility of the proposed model is verified through case studies. The results indicate that the model can objectively and accurately evaluate the credit levels of electricity sales companies.