赵会茹, 李兵抗, 苏群, 张圆圆, 薛万磊. 基于博弈论组合赋权和改进TOPSIS的新能源发电商信用风险评价模型研究[J]. 现代电力, 2023, 40(4): 514-524. DOI: 10.19725/j.cnki.1007-2322.2022.0027
引用本文: 赵会茹, 李兵抗, 苏群, 张圆圆, 薛万磊. 基于博弈论组合赋权和改进TOPSIS的新能源发电商信用风险评价模型研究[J]. 现代电力, 2023, 40(4): 514-524. DOI: 10.19725/j.cnki.1007-2322.2022.0027
ZHAO Huiru, LI Bingkang, SU Qun, ZHANG Yuanyuan, XUE Wanlei. Research on Credit Risk Evaluation Model of New Energy Power Producers Based on Game Theory Combination Weights and Improved TOPSIS[J]. Modern Electric Power, 2023, 40(4): 514-524. DOI: 10.19725/j.cnki.1007-2322.2022.0027
Citation: ZHAO Huiru, LI Bingkang, SU Qun, ZHANG Yuanyuan, XUE Wanlei. Research on Credit Risk Evaluation Model of New Energy Power Producers Based on Game Theory Combination Weights and Improved TOPSIS[J]. Modern Electric Power, 2023, 40(4): 514-524. DOI: 10.19725/j.cnki.1007-2322.2022.0027

基于博弈论组合赋权和改进TOPSIS的新能源发电商信用风险评价模型研究

Research on Credit Risk Evaluation Model of New Energy Power Producers Based on Game Theory Combination Weights and Improved TOPSIS

  • 摘要: 新能源发电商参与市场是实现我国能源体系清洁低碳转型的关键路径,但其自身出力的不确定性会引发信用风险,因此有必要对其信用风险进行评价,为推进企业信用风险分类管理提供参考。基于意愿—能力—行为框架设计了包含3个一级指标14个二级指标的新能源发电商信用风险评价指标体系,构建了基于博弈论组合反熵权法和分层赋权法的指标赋权方法,以及基于差商灰色关联改进TOPSIS法的信用风险评价方法,以7个新能源发电商为例进行了分析。结果表明,合同电量履约率、现货市场申报偏差率、出力预测偏差率是影响新能源发电商信用风险的关键因素,此外主体的不正当竞争行为和偷税漏税行为也是市场信用管理机构需要关注的重点。排序一致性检验和留一法(Leave-One-Out,LOO)分析表明所提出的模型具有较强的鲁棒性。

     

    Abstract: The participation of new energy power producers (abbr. NEPP) in the market is the key path to realize the clean and low-carbon transformation of China’s energy system, but the uncertainty of its own output will lead to credit risk. therefore, it is necessary to evaluate NEPP’s credit risk and provide reference for promoting the classified management of enterprise credit risk. Based on the aspiration-ability-action framework, a credit risk evaluation index system of NEPP, in which three primary indices and 14 secondary indices were included, was designed. An index weighting method based on game theory combined with anti-entropy-weight (abbr. AEW) method and level based weight assessment (LBWA) method, and hierarchical weighting method, and a credit risk evaluation method based on difference and quotient grey correlation analysis and the credit risk evaluation method based on difference and quotient grey relation analysis improved technique for order preference by similarity to an ideal solution (DQGRA-TOPSIS) method were constructed. Seven NEPPs were analyzed as the example, and the results show that the contract electricity compliance rate, spot market declaration deviation rate and output prediction deviation rate are the key factors affecting the credit risk of NEEPs. In addition, unfair competition of NEEP and the tax evasion behavior are also the focus of market credit management institutions. Results of ranking consistency test and LOO (Leave-One-Out) analysis show that the proposed model possesses strong robustness.

     

/

返回文章
返回