Microgrid frequency control method based on reinforcement learning

Abstract

The invention provides a microgrid frequency control method based on reinforcement learning, and the method comprises the steps: 1, building a microgrid frequency control model: controlling the outputpower in a mode of controlling the frequency and voltage amplitude, and adjusting a droop parameter for frequency modulation control; step 2, a frequency control strategy based on reinforcement learning: designing a state space, an action set and a reward function according to a specified frequency, and training a Q table to select an optimal correction amount; and step 3, establishing a microgrid frequency control system based on reinforcement learning: training different scenes based on Q learning, and verifying the effectiveness and adaptability of the proposed method from multiple aspectsof a learning training process and frequency control response characteristics.

Description

technical field [0001] The invention relates to the field of micro-grid frequency control, in particular to a micro-grid frequency control method based on reinforcement learning. Background technique [0002] The microgrid combines energy storage devices, loads, and power generation units to form a controllable unit, and connects distributed power sources to the grid in the form of micro grids to achieve full utilization of distributed power sources. Microgrid is an important technical way to realize new energy grid connection, improve its power supply quality, and increase the utilization rate of new energy. [0003] Different from the traditional generator set, the distributed power connected to the microgrid does not have rotor inertia, which leads to a low equivalent inertia of the microgrid. In the grid-connected mode, the main grid can also be used for frequency support, while in the isolated grid mode , due to the low inertia of the system, once a load fluctuation oc...

AI Technical Summary

Problems solved by technology

[0003] Different from the traditional generator set, the distributed power connected to the microgrid does not have rotor inertia, which leads to a low equivalent inertia of the microgrid. In the grid-connected mode, the main grid can also be used for frequency support, while in the isolated grid mode , due to the low inertia of the system, once the load fluctuates, it will have a great impact on the system frequency, and even cause a power outage

And because most of the power generation units in the microgrid are controlled by droop control method, when the microgrid is running in isolation, its operating frequency can easily deviate from the operating frequency of the power system stipulated by the state.

Images