Multi-agent power generation optimization scheduling method based on reinforcement learning

Abstract

The invention provides a multi-agent power generation optimization scheduling method based on reinforcement learning. With the goal of maximizing the total operation benefit, a multi-agent complementary optimization model is established; based on the established multi-agent complementary optimization model, a multi-agent game is constructed. Model, according to the Nash game theory, the local optimal strategy under the mutual coordination of each agent is obtained, and the local optimal strategy set is constructed; the Q-learning algorithm is used to solve the optimization problem, and the global optimal is obtained, which is the optimal strategy set π * According to the idea of ​​combining Nash game and Q-learning algorithm, this method can transform the multi-agent optimization problem of complex system into a state-action value function convergence problem, and obtain the best optimization scheme, reduce the complexity of optimization scheduling, and reduce the complexity of optimization scheduling. It can realize model-free optimization and achieve the purpose of energy saving and high efficiency.

Description

technical field [0001] The invention relates to a multi-agent power generation optimization scheduling method based on reinforcement learning. Background technique [0002] Due to the large number of distributed energy sources involved in the multi-agent system and the strong randomness between the power generation side and the load side, the optimal scheduling of multi-agent systems presents the characteristics of multiple constraints and strong uncertainty. The traditional centralized optimization scheduling method has the problem of unsatisfactory effect. [0003] The above problems should be considered and solved in the multi-agent optimal scheduling process. SUMMARY OF THE INVENTION [0004] The purpose of the present invention is to provide a multi-agent power generation optimization scheduling method based on reinforcement learning to solve the problem of unsatisfactory effect in the traditional centralized optimal scheduling method in the prior art. [0005] The ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a multi-agent power generation optimization scheduling method based on reinforcement learning to solve the problem of unsatisfactory effects in the traditional centralized optimization scheduling method in the prior art

Images