Document Type

Article

Publication Date

2026

DOI

10.1109/ACCESS.2026.3735618

Publication Title

IEEE Access

Volume

14

Pages

164629-164643

Abstract

Low-frequency oscillations remain a major concern in modern power systems with increasing penetration of renewable energy sources. Conventional damping methods, such as power system stabilizers and fixed-gain controllers, may not provide sufficient damping for converter-interfaced wind generators under changing operating conditions and disturbances. To address these limitations, this paper proposes a deep reinforcement-learning-enhanced model predictive control (DRL-MPC) framework for damping inter-area oscillations in multi-machine power systems incorporating permanent magnet synchronous generator (PMSG)-based wind turbines. The proposed approach leverages the energy stored in the DC-link capacitor as a fast dynamic buffer to enable supplementary active-power modulation for oscillation damping. A twin-delayed deep deterministic policy gradient (TD3) algorithm is integrated with MPC to adaptively tune the MPC weighting parameters online to enhance damping performance under varying operating conditions. The proposed DRL-MPC controller is validated in MATLAB/Simulink on a 100-MVA modified IEEE 9-bus system incorporating a PMSG-based wind farm and three interconnected generator subsystems under three disturbance scenarios: a three-phase fault, a load change, and a wind-speed change. The simulations assume accurate measurements and negligible communication delays. Simulation results demonstrate that the proposed method outperforms conventional MPC and H-infinity controllers by providing faster damping and improved stability. Quantitative performance metrics, including MSE, MAE, and RMSE, validate the effectiveness of the proposed approach in reducing tracking errors and improving damping performance. The results confirm that integrating data-driven learning with predictive control enables adaptive MPC weight tuning, faster oscillation damping, and effective DC-link energy utilization for damping support.

Rights

© 2026 The Authors.

This work is licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) License.

Original Publication Citation

Elhaji, E., Liu, S., Chaoui, H., & Egbomwan, O. (2026). Deep reinforcement learning-enhanced model predictive control using DC-link energy modulation for inter-area oscillation damping in PMSG-based power systems. IEEE Access, 14, 164629-164643. https://doi.org/10.1109/ACCESS.2026.3735618

ORCID

0000-0001-8728-3653 (Chaoui)

Download

Share

COinS