| Since the promulgation of the “531” New Policy,the replacement of the current FIT policy by the RPS policy has gradually become the main direction of the reform of the renewable energy policy system.In 2021,the renewable energy quota system will be implemented nationwide.The impact of the energy industry.This article focuses on the micro perspective,considering how to choose the optimal investment time and optimal installed capacity of photovoltaic power generation projects under the background of the implementation of the quota system,and consider the sequential decision-making problem of phased investment behavior.The main contents of this paper are as follows:This article first summarizes the theoretical basis and domestic and foreign literature research.According to the research object and purpose,the real option theory and the reinforcement learning theory are determined as the research foundation of this article,and then the domestic and foreign renewable energy policy and power market simulation research Relevant literature is summarized,analyzed,and reviewed to establish a literature basis.By reviewing the existing literature at home and abroad,this article found that the current policy background in the research on investment decisions of renewable energy companies is mostly related to government subsidies,feed-in tariffs,and carbon taxes,and there is a lack of research on the impact of the quota system and the green certificate market.At the same time,there is also a lack of research on the renewable energy quota system from a micro-level perspective.Then,the characteristics of photovoltaic power generation projects were sorted out from the four aspects of investment concentration,cost change expectations,policy impacts,and natural environmental impacts,and then clarified that the investment decision-making issues of related projects under the background of the implementation of the renewable energy quota system include investment irreversibility.The features of option,including the nature,risk and flexibility of decision-making,determine the feasibility of the real option method to solve the problem.Based on the above foundation,the real option method is used to establish a mathematical model,and the risk-neutral analysis method is used to quantitatively analyze the uncertain factors.Finally,an analytical solution for the optimal investment time and investment scale of a single decision-making process is obtained,and multistage sequential investment Decision-making problems are solved by computer simulation.In the experiment,the kernel function-based reinforcement learning algorithm was used to simulate the actual decision-making process,and the optimal investment decision-making scheme of a photovoltaic power generation project under the background of the implementation of the quota system was studied.The specific modeling scheme of numerical simulation was first proposed,and then selected according to the literature.Determine the core parameters,clarify the power generation company’s state and action feasible region on this basis,use the reinforcement learning method to simulate specific examples,and evaluate and analyze the results of the simulation model.The research results show that staged investment can significantly increase the flexibility of investment,help investors avoid risks,and actually increase the value of the entire project.In addition,research on technical costs and green certificate price levels shows that the reduction of technical costs is conducive to The advance deployment of investment projects,and the increase in the price level of green certificates also promoted the arrival of the optimal investment time in advance.The research results of this article will contribute to the actual quality of investment decision-making,and at the same time provide a micro-level research perspective for the policy impact of the quota system. |