Optimizing intelligent startup strategy of power system using PPO algorithm.
Saved in:
| Title: | Optimizing intelligent startup strategy of power system using PPO algorithm. |
|---|---|
| Authors: | Sun, Yan, Wu, Yin, Wu, Yan, Liang, Kai, Dong, Cuihong, Liu, Shiwei |
| Source: | Intelligent Decision Technologies; 2024, Vol. 18 Issue 4, p3091-3104, 14p |
| Subject Terms: | MACHINE learning, ELECTRIC lines, MATHEMATICAL optimization, RENEWABLE energy sources, DYNAMIC models |
| Abstract: | This article aimed to use the proximal policy optimization (PPO) algorithm to address the limitations of power system startup strategies, to enhance the adaptability, coping ability, and overall robustness of the system to variable grid demand and integrated renewable energy, the constraints in the power system start-up strategy are optimized. Firstly, this article constructed a dynamic model of the power system, including key components such as generators, transformers, and transmission lines; secondly, it integrated the PPO algorithm and designed interfaces that allow the algorithm to interact with the power system model; afterward, the state variables of the power system were determined, and a reward function was designed to evaluate the startup efficiency and stability of the system. Next, this article adjusted the reward function and trained and iterated multiple times in the simulation environment to guide the algorithm to learn the optimal startup strategy. Finally, an effective evaluation of the strategy can be conducted. The research results showed that after optimization by the PPO algorithm, the stable frequency startup of the power system only took about 23 seconds, and the system recovery time was reduced by 33.3% under a sudden load increase. The algorithm used can significantly optimize the intelligent startup strategy of the power system. [ABSTRACT FROM AUTHOR] |
| Copyright of Intelligent Decision Technologies is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Complementary Index |
Nájsť tento článok vo Web of Science Be the first to leave a comment!