Method and equipment for determining medium-and-long-term electricity market transaction strategy of wind power plant

The invention discloses a method and equipment for determining a long-term electricity market transaction strategy in a wind power plant, and the method comprises the steps: carrying out the mathematical modeling of uncertain factors through a scene generation method after the uncertain factors of a distributed wind power plant are divided according to seasons, months and weeks, and obtaining a training set and a test set; mathematical modeling is performed according to historical data of related facilities of the distributed wind power plant divided by seasons, months and weeks, a reinforcement learning process including a virtual environment is formed, and the related facilities include operation states of a current collection circuit, a box-type substation and a fan; training an agent according to the training set and the virtual environment, and obtaining a depth enhancement algorithm model which is obtained based on the purpose of maximizing an accumulated value of a value function of the agent; and test