Voltage Decentralized Real-Time Control Considering Monotonicity Constrained Deep Deterministic Policy Gradient Algorithm

View at publisher Summarise with VIRAT
Authors: Jiang Changjun, Liu Haitao, Xiong Xiong, Ji Yu, Xu Yini, Ran Xi
Cited by: 0
References: 4

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Deep reinforcement learning (DRL) serves as a crucial method for addressing real-time control in power systems. However, its practical application is limited due to the lack of explicit stability analysis. This paper proposes a real-time voltage control strategy based on a monotonicity-constrained Deep Deterministic Policy Gradient (DDPG) algorithm. By constructing a Lyapunov function, explicit control stability constraints are established. The strategy parameterizes each local voltage controller using monotonic neural networks, integrating this stability-constrained architecture with DRL to ensure the algorithm satisfies monotonicity constraints. We demonstrate the effectiveness of our approach in both single-phase and three-phase IEEE test feeders, where the proposed method can shorten the voltage recovery time by 80.82% compared to traditional DDPG algorithm. the voltage recovery is stabilized in all test scenarios, The algorithm without monotonicity constraints cannot guarantee 100% voltage stability control.

Fields of Study & Subjects

Optimal Power Flow Distribution Microgrid Control and Optimization Power System Optimization and Stability

Publication details

Journal / source
IEEE PES International Meeting (PES IM)
Pages
1–5
Publication date
18 January 2026
Language
EN