Optimal Investment Strategy for Wind Power under Electricity-Carbon-Green Certificate Trading: Based on Multi-Agent Deep Reinforcement Learning

The carbon emission trading market and the trading green certificate market promote the realization of carbon reduction targets, and they are coupled with the electricity market. This paper analyzes the market equilibrium of wind power investment strategy under these three markets. The wind power output samples are obtained by Latin hypercube sampling, and the investors take into account the risk preference represented by Conditional Value at Risk in addition to the average income. The multi-agent deep reinforcement learning method is applied to solve the equilibrium of wind power investment strategy. The example shows that the algorithm is convergent and the investment strategy equilibrium can be found. Moreover, the wind power investment under different carbon reduction efforts and different renewable energy consumption responsibilities is compared.