Park microgrid load optimization scheduling method and system based on two-stage reinforcement learning

Abstract

The invention provides a park microgrid load optimization scheduling method and system based on two-stage reinforcement learning, and relates to the technical field of microgrid load optimization scheduling. The method comprises the following steps: acquiring relevant parameters of each microgrid in a park, and then acquiring the optimal price of each microgrid by a load agent based on the relevant parameters by using a reinforcement learning algorithm based on a random strategy gradient; and finally, based on the optimal price, performing optimization scheduling on each microgrid of the park by using a deep reinforcement learning Actor-Critic algorithm. According to the technical scheme, the load agent participates in, two-stage reinforcement learning is used as an optimal price acquisition and micro-grid optimization scheduling algorithm, and under the conditions that incomplete user information and environment information are acquired and specific operation models and parameters of equipment are not depended on, the optimal strategy for the operation of the microgrid in the park can be provided more accurately, timely and efficiently; meanwhile, the privacy of microgrid users can be protected and the economical efficiency of the microgrid can be improved.

Description

technical field [0001] The invention relates to the technical field of microgrid load optimization scheduling, in particular to a method and system for park microgrid load optimization scheduling based on two-stage reinforcement learning. Background technique [0002] In order to adapt to the development strategy of energy system transformation, the park energy micro-grid has become an important role in regional energy consumption. With the development of the park micro-grid system, there are often multiple park micro-grid energy systems in the same power distribution park. The continuous deepening of power market reform provides new opportunities for the grid-connected operation of the park micro-grid. With the opening of the power market, the microgrid in the park will be able to participate in regional power dispatching in an independent capacity, and improve the distribution efficiency of power loads through two-way interaction with power distribution companies. If the ...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the existing technology, the present invention provides a two-stage reinforcement learning-based load optimization scheduling method and system of the park microgrid, which solves the problem that the existing technology cannot accurately obtain the park microgrid load when the user information and environment information are incomplete or changed. Precise and Efficient Load Optimal Scheduling of Microgrid

Images