Research On The Credit Risk Measurement And Control Model Of Multi Power Market Entities

Since 2019,China has issued intensive policy documents to promote the construction of power market,accelerated the integration of power market scope and market model,and deepened the construction of high-standard power market system.It is of great significance to support the construction of national unified power market,implement the new energy security strategy of "four revolutions and one cooperation" proposed by General Secretary Xi Jinping and achieve the carbon peaking and carbon neutrality goals.The power market has the characteristics of complex system,and the interests of different market entities are intertwined in the market transaction,which intensifies the self-interest,complexity and uncertainty of the trading behavior of market entities.In 2015,the policy document on deepening the reform of power system proposed to establish and improve the credit system of market entities and standardize the market order.On January 4,2022,Premier Li Keqiang presided over the executive meeting of the State Council and proposed that credit is the basis of market economy and that classified management of enterprise credit risk should be promoted in accordance with laws and regulations.In this context,to promote the construction of high-standard power market system and national unified power market,it is necessary to establish a perfect credit system and risk management measures on the premise of meeting the safe and stable operation of power grid,so as to restrict the entity’s trading behavior and standardize the trading order of power market.The trading behavior of the power market entity is dynamic,and the measurement,prevention and control of the credit risk of multiple entities are also a dynamic process.The government,industry and enterprises have carried out a lot of research and application work on the credit system of the power market,but they have not fully considered the trading behavior characteristics of the market entities,and the reflection on the credit level and risk status of the market entities is not comprehensive enough.Hence,it is difficult to restrict the entities’ unexpected behavior and prevent and control the credit risk caused by the behavior.Therefore,it is of great theoretical significance and practical value to examine the credit risk measurement and risk prevention and control of multiple entities in power market,considering the trading behavior characteristics of power market entities as the breakthrough point.In view of this,based on the core content of "credit risk measurement and risk prevention and control”in the power market credit system construction,this paper carries out the construction of multi entities trading behavior deduction model in the power market,multi entities credit rating model considering trading behavior,multi entities credit risk measurement model considering credit rating,and multi entities credit risk prevention and control model based on behavior-credit-risk transmission,which can provide support for standardizing the trading behavior of multi entities in the market and reducing credit risk.The main research contents of this paper are as follows:(1)A multi entities trading behavior deduction model in power market considering market connection and adaptive learning game mechanism is constructed,and the trading behavior characteristics of multi entities are examined accordingly.Firstly,the connection relationship of power market in different time scales and the main forms of trading behavior of market entities in different markets are analyzed.Secondly,taking the medium and long-term contract electricity decomposition as the representation of the market connection relationship,and introducing the adaptive learning game mechanism,this paper constructs a multi entities trading behavior deduction model.Thirdly,considering the bilateral uncertainty of supply and demand in the model,a stochastic hierarchical scenario generation method is proposed,the trading behavior deduction model is reconstructed into a two-stage hybrid stochastic optimization problem,and a model solution process based on AGOHSA method is proposed.Finally,an example simulation is carried out in a system including 10 conventional generators,7 new energy generators and 15 electricity purchasers.The results show that when the market supply and demand basically match,the trading behavior deduction model can better simulate the process of spontaneous competition and market equilibrium,and under the equilibrium conditions,there are significant differences in the trading behavior characteristic indicators such as the proportion of medium and long-term contract electricity,the performance rate of contract electricity and the deviation rate of spot declaration.(2)This paper reveals the influence mechanism of trading behavior on the entity’s credit,and constructs the market entity’s credit rating index system and model considering trading behavior.Firstly,this paper constructs the transmission model of multi entities trading behavior on entities credit based on Grey-DEMATEL-ISM,and reveals the influence mechanism of entity’s trading behavior on credit.Secondly,based on the 3A framework,this paper constructs the credit rating index system of multiple entities in the power market from the three dimensions of performance aspiration,performance ability and performance action.Thirdly,the multi entities credit rating model based on hybrid multi-criteria decision-making is constructed,including the index weighting method based on the combination of AEW and LBWA based on game theory,and the credit rating method based on DQGRA-MARCOS.Finally,the empirical analysis of conventional generators,new energy generators and electricity purchasers is carried out.The results show that the key factors affecting the credit rating of different types of market entities are different,but the performance rate of contract electricity,unfair competition,output/load forecast deviation,market-oriented transaction scale,and tax evasion have a great impact on the credit rating of the market entities.Compared with single weighting method,conventional MARCOS and TOPSIS models,the results show that the model constructed in this paper performs well in ranking consistency,reverse order and sample separation,and can better improve the decision-making efficiency on the premise of ensuring robustness.(3)The traditional Credit Metrics model is improved,and a suitable credit risk measurement model for multiple entities in the power market is constructed.Based on the theoretical framework of Credit Metrics model,this paper puts forward the construction method of market multi entities credit rating transfer matrix,and constructs the multi entities forward loss calculation model based on performance guarantee,and the multi entities credit risk value calculation model based on conditional risk.After orderly combination,the multi entities credit risk measurement model in power market considering credit rating is formed.Through the analysis of examples,the credit risk values of different market entities are measured quantitatively.The results show that the credit rating transfer of the market entity has a certain inertia,and the credit rating of market entities in the next period will be the same as that in the previous period in a great probability.If the credit rating is transferred,the probability of transferring to the adjacent rating is greater than that of transferring to other ratings.Under the same confidence level,the CVaR value of market entity is significantly greater than that of VaR,that is,using VAR will underestimate the credit risk level to a certain extent.Therefore,it is more credible to measure the credit risk of the market entity by CVaR method.(4)This paper constructs the system dynamics model for the behavior-credit-risk transmission of market entity,and puts forward targeted measures and suggestions on market entity’s credit risk prevention and control.Firstly,considering the interaction relationship among market entity’s trading behavior,credit rating and credit risk,a market entity’behavior-credit-risk transmission model including trading behavior subsystem,credit rating subsystem and credit risk subsystem is constructed for conventional generator,new energy generator and electricity purchaser by using system dynamics method.Secondly,the validity of the model is verified by structure test and behavior reproduction ability test,and the simulation analysis is carried out for different market entities.Finally,according to the simulation analysis results,this paper puts forward targeted measures and suggestions on credit risk prevention and control of market entities.The results show that marginal cost,output forecast deviation rate and load forecast deviation rate are the key impact variables in the behavior-credit-risk transmission system of conventional generator,new energy generator and electricity purchaser respectively.The changes of these variables can significantly affect the trading behavior characteristic indicators and cost-benefit of the entity in the market,and then affect the credit rating results and credit risk value of the entity.However,due to the non-linear correspondence between the credit rating comprehensive utility function value and credit rating,the impacts of thses variables on credit risk value are weakened.Considering the characteristics of trading behavior,this paper examines the credit rating,credit risk measurement and credit risk prevention and control of multiple entities in the power market,reflecting the practical needs of China’s power market construction and enterprise credit risk classification management.It has important theoretical value and practical significance for scientifically grasping the credit level and risk status of multiple entities in the power market and effectively preventing and controlling the credit risk of entities.The deduction model of multi entities trading behavior constructed in this paper extends the relevant research on multi entities trading strategy in the complex environment of power market,and the constructed entity credit rating model,entity credit risk measurement model and behavior-credit-risk transmission model enrich the relevant achievements of entity credit risk management in the power market.These achievements can effectively support the modernization of power market operation and governance system and capacity,help the construction of high-standard power market system and the establishment and improvement of enterprise credit classification management mechanism,and also have a certain reference value for credit management in other markets.