Wind power dispatching method and system based on improved NishQ algorithm

Abstract

The invention relates to a wind power dispatching method and system based on an improved NishQ algorithm. The method comprises the steps of S1, building dispatching models and operation constraint conditions of a coal-fired thermal power generating unit side, an energy storage facility side, an electric vehicle battery swap station side and a wind power plant side in a competitive power market, and taking the maximization of the operation benefit of each main body as a target, and constructing a multi-agent Nash equilibrium game model to find an optimal wind power scheduling strategy; S2, solving a multi-agent game problem by adopting a NishQ algorithm, using the JS divergence for comparing own historical experience with the Nash equilibrium state, optimizing the learning rate of each agent at different moments, improving the Nash Q algorithm, and shortening the time required by the optimal strategy of each agent, and S3, building a combined operation model of the wind power plant and the electric vehicle battery swap station as well as the wind power plant and the energy storage station, and solving the economic benefit of the combined operation model by adopting an improved NishQ algorithm under the market game. The method and the system are favorable for optimizing the dispatching effect of the wind power plant.

Description

technical field [0001] The invention belongs to the technical field of power system optimization operation and scheduling, and in particular relates to a wind power scheduling method and system based on an improved NashQ algorithm. Background technique [0002] With the goal of "carbon peak" and "carbon neutrality", in order to realize the transformation of my country's energy structure, the installed capacity of wind turbines will gradually increase. Grid integration presents great challenges. [0003] At present, there are two main types of research on wind power optimal dispatching: one is to set the objective function from the perspective of wind power operation benefits, and establish an optimal dispatching model to optimize wind power operation; The operation target is set from the perspective of power grid operation efficiency, so as to optimize the wind power dispatching strategy. [0004] However, because the power generation facilities in the micro-grid are constr...

AI Technical Summary

Problems solved by technology

[0004] However, since the power generation facilities in the micro-grid are constructed by different investment entities, the micro-grid is dispatched as a whole, which fails to fully consider the benefits of each investment entity, which is not conducive to mobilizing the enthusiasm of all parties.

The literature on optimizing wind power dispatching strategies from the perspective of wind power only considers the maximization of the benefits of wind farm investors, ignores the benefits of other investment entities, and cannot reflect the real benefits of wind power after entering the power market. From a long-term perspective, it is not conducive to wind power. sustainable development after entering the competitive power market

[0005] Existing research uses multi-agent reinforcement learning algorithm to solve multi-agent sequential decision-making problems, but the learning of agents is often limited to their own learning experience and is not sensitive to the external environment, which makes the convergence speed slow. incurs a large computational cost in

Images