Adaptive doubly-fed fan subsynchronous oscillation additional damping control method based on knowledge and deep reinforcement learning
Description
As global demand for clean energy continues to rise, wind energy has emerged as a crucial contributor to electricity production due to its abundant renewable nature. However, the expansion of wind power generation systems, particularly those employing doubly-fed induction generator (DFIG) units and series capacitor compensation lines, presents challenges, notably in mitigating sub-synchronous oscillations (SSOs). These oscillations adversely affect the system's frequency and voltage stability. This study proposes an innovative supplementary damping control method tailored to address SSO issues, ensuring the safe and stable operation of wind power grid-connected systems. This study introduces an intelligent damping controller that integrates the improved Twin Delayed Deep Deterministic Policy Gradients (TD3) algorithm. It reframes the SSO suppression problem as a Markov Decision Process (MDP) and utilizes an enhanced TD3 algorithm to solve it. By incorporating the softmax operation, the value function underestimation bias inherent in the TD3 algorithm is mitigated, enhancing the damping control process's efficiency. Additionally, a knowledge fusion strategy is employed to accelerate training convergence and improve the suppression effectiveness of the intelligent agent, overcoming the low training efficiency issue of current purely data-driven methods. Experimental results illustrate that the proposed Intelligent Sub-Synchronous Damping Controller (I-SSDC) effectively suppresses SSO across various operating conditions, demonstrating robust dynamic tracking control capabilities. Compared to traditional sub-synchronous damping controllers (T-SSDC), the I-SSDC exhibits superior adaptability and robustness across a broader operating range, maintaining effectiveness even without training samples. The deep reinforcement learning-based approach introduced in this study offers a novel solution for mitigating sub-synchronous oscillations in wind power systems, with significant theoretical and practical implications for enhancing the stability and reliability of wind power systems.