Robust Multi-Layered Control of Shunt Hybrid Active Power Filter Using HQC, ASMC and MPC

Abstract

This paper presents a novel robust control framework for a Shunt Hybrid Active Power Filter (SHAPF) to mitigate harmonics, compensate reactive power, and ensure dynamic stability in distribution networks with nonlinear loads. The proposed methodology integrates three advanced control strategies. First, a half-quadratic criterion–based (HQC) adaptive filtering algorithm is employed for reference current generation, enhancing robustness against impulsive and non-Gaussian disturbances while achieving faster convergence than conventional LMS/LMF-based filters. Second, an adaptive sliding mode controller (ASMC) is designed for DC-link voltage regulation, offering robustness to parameter uncertainties and rapid transient recovery without excessive chattering, thereby outperforming traditional PI controllers. Third, a finite control set model predictive current control (FCS-MPC) approach is implemented for the current loop, enabling direct switching state optimization to minimize harmonic distortion and improve dynamic response. Simulation results in MATLAB/Simulink confirm the superiority of the proposed strategy, with greater THD reduction under steady-state and during transients, improved DC-link stability under sudden load variations, and balanced source currents during unbalanced loading. Compared to conventional adaptive filters, PI-controlled DC-link regulation, and hysteresis current controllers, the proposed hybrid control demonstrates superior robustness, adaptability, and computational efficiency, making it an effective solution for power quality problems.