A reinforcement learning scheduling method for power system based on data-driven pretraining

Hao Zhang; Tianbo Zhu; Du Xu; Qingfeng Tang; Huan Luo

doi:10.1117/12.3024478

13 May 2024 A reinforcement learning scheduling method for power system based on data-driven pretraining

Hao Zhang, Tianbo Zhu, Du Xu, Qingfeng Tang, Huan Luo

Proceedings Volume 13159, Eighth International Conference on Energy System, Electricity, and Power (ESEP 2023); 131598S (2024) https://doi.org/10.1117/12.3024478
Event: Eighth International Conference on Energy System, Electricity and Power (ESEP 2023), 2023, Wuhan, China

Abstract

With the fast development of renewable energy, a large amount of renewable energy is integrated into the power system. However, the intermittency and volatility of renewable energy sources such as solar and wind energy may pose huge challenges to power system scheduling. In order to reduce the impact of renewable energy volatility on the operation of the power system and improve the autonomy of the power system. This paper proposes a reinforcement learning scheduling method for power system based on data-driven pretraining. Firstly, it utilizes ge2e to encode wind, photovoltaic, and load, and performs data-driven pre training to obtain the embedding vectors of different sourcesThen, the embedding vectors are used as the input features of the state, and the power system is scheduling based on the reinforcement learning method-SAC. This method is able to keep secure real-time scheduling of power system with highly volatile loads and renewable energy.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Hao Zhang, Tianbo Zhu, Du Xu, Qingfeng Tang, and Huan Luo "A reinforcement learning scheduling method for power system based on data-driven pretraining", Proc. SPIE 13159, Eighth International Conference on Energy System, Electricity, and Power (ESEP 2023), 131598S (13 May 2024); https://doi.org/10.1117/12.3024478

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available