Adaptive Active Disturbance Rejection Load Frequency Control for Power System with Renewable Energies Using the Lyapunov Reward-Based Twin Delayed Deep Deterministic Policy Gradient Algorithm

Yuemin Zheng, Jin Tao (), Qinglin Sun (), Hao Sun, Zengqiang Chen and Mingwei Sun
Additional contact information
Yuemin Zheng: College of Artificial Intelligence, Nankai University, Tianjin 300350, China
Jin Tao: Silo AI, 00100 Helsinki, Finland
Qinglin Sun: College of Artificial Intelligence, Nankai University, Tianjin 300350, China
Hao Sun: College of Artificial Intelligence, Nankai University, Tianjin 300350, China
Zengqiang Chen: College of Artificial Intelligence, Nankai University, Tianjin 300350, China
Mingwei Sun: College of Artificial Intelligence, Nankai University, Tianjin 300350, China

Sustainability, 2023, vol. 15, issue 19, 1-25

Abstract: The substitution of renewable energy sources (RESs) for conventional fossil fuels in electricity generation is essential in addressing environmental pollution and resource depletion. However, the integration of RESs in the load frequency control (LFC) of power systems can have a negative impact on frequency deviation response, resulting in a decline in power quality. Moreover, load disturbances can also affect the stability of frequency deviation. Hence, this paper presents an LFC method that utilizes the Lyapunov reward-based twin delayed deep deterministic policy gradient (LTD3) algorithm to optimize the linear active disturbance rejection control (LADRC). With the advantages of being model-free and mitigating unknown disturbances, LADRC can regulate load disturbances and renewable energy deviations. Additionally, the LTD3 algorithm, based on the Lyapunov reward function, is employed to optimize controller parameters in real-time, resulting in enhanced control performance. Finally, the LADRC-LTD3 is evaluated using a power system containing two areas, comprising thermal, hydro, and gas power plants in each area, as well as RESs such as a noise-based wind turbine and photovoltaic (PV) system. A comparative analysis is conducted between the performance of the proposed controller and other control techniques, such as integral controller (IC), fractional-order proportional integral derivative (FOPID) controller, I-TD, ID-T, and TD3-optimized LADRC. The results indicate that the proposed method effectively addresses the LFC problem.

Keywords: load frequency control; Lyapunov reward; twin delayed deep deterministic policy gradient algorithm; linear active disturbance rejection control; renewable energy sources (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/15/19/14452/pdf (application/pdf)
https://www.mdpi.com/2071-1050/15/19/14452/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:15:y:2023:i:19:p:14452-:d:1253033

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().