Research on Dynamic Self-Tuning Method of Control Parameters for AVC Substations in Wind Farm Clusters

In wind farm clusters, automatic voltage control (AVC) substations frequently experience voltage regulation oscillation or delay under no-communication conditions, primarily due to voltage-reactive power sensitivity mismatch among distributed wind farms. To address this critical issue, this study proposes a voltage dynamic response-based control parameter self-tuning method that operates entirely without inter-substation communication. First, the reactive power coupling mechanism among wind farms within a cluster is systematically analyzed, revealing that fixed sensitivity parameters are inherently incapable of ensuring stable voltage regulation under dynamically varying operating conditions. Second, a reactive power step correction coefficient is introduced to formulate a comprehensive AVC control parameter, and a sensitivity mismatch criterion based on both voltage variation trend and regulation time is constructed to enable online identification of under-regulation and over-regulation conditions. On this basis, two self-tuning strategies-fixed-step and variable-step-are developed, with the variable-step method adaptively adjusting the step size according to real-time voltage deviation characteristics to achieve faster convergence. Finally, comprehensive simulation results demonstrate that the proposed method effectively suppresses voltage oscillations and significantly reduces regulation time without requiring any communication infrastructure. Under typical mismatch scenarios, the voltage is able to enter the acceptable deadband within two adjustment steps, confirming the effectiveness and practical applicability of the proposed approach for enhancing voltage stability in large-scale wind farm clusters.