Indirect Adaptive Polynomial Wavelet-Based Neuro-Fuzzy Controller for STATCOM-Equipped Power Systems

Abstract

Modern power systems face growing stability challenges due to rising network complexity and dynamic operating conditions. Traditional control mechanisms often struggle to effectively mitigate Low-Frequency Oscillations (LFOs), underscoring the need for more advanced and adaptive damping strategies. Flexible AC Transmission Systems (FACTS), especially Static Synchronous Compensators (STATCOMs), have shown considerable promise in strengthening system stability under such challenging conditions. However, their performance is highly dependent on the quality of the Supplementary Damping Controller (SDC) strategy, and conventional methods may fall short under nonlinear and dynamic conditions. To tackle these issues, this paper presents a novel Indirect Adaptive Polynomial Wavelet-based Neuro-Fuzzy Control (ANFWC) framework designed to damp LFOs in STATCOM applications. The ANFWC includes three controllers, each employing a distinct Orthogonal Polynomial Wavelet-based Neural Network (PWNN) within an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based Takagi-Sugeno-Kang (TSK) controller: the Legendre Wavelet-based Controller (ANFLWC), the Hermite Wavelet-based Controller (ANFHWC), and the Chebyshev Wavelet-based Controller (ANFCWC). These controllers enhance ANFIS learning and nonlinear mapping by leveraging PWNNs in the consequent layer. The performance of these controllers is evaluated through MATLAB simulations on the Single-Machine Infinite Bus (SMIB) and IEEE 9-bus Western System Coordinating Council (WSCC) test systems under various fault and disturbance conditions. Comparative analyses show that ANFLWC achieves the best performance, followed by ANFCWC and ANFHWC. All proposed controllers significantly outperform the conventional ANFIS-based TSK controller (ANFTSKC) and Lead-Lag Control (LLC), demonstrating the effectiveness of the ANFWC approach in improving power system damping and stability.