Robust Intelligent Control of Active Power Filter Using Neural Learning Mechanism

The problem of harmonic distortion becomes more and more serious with the extensive application of power electronic equipments. This paper proposes a meta-cognitive fuzzy-neural-network controller based on second-order sliding mode for active power filter (APF) to suppress harmonics. First, the method of second-order sliding mode control (SOSMC) is used to keep tracking of the APF reference current. Next, the meta-cognitive fuzzy-neural-network (MCFNN) with exact approximation ability is further studied to approximate the unknown part in SOSMC. Subsequently, the stability analysis is carried out through Lyapunov theory to ensure the tracking performance of the closed-loop system. Finally, simulation studies have demonstrated the excellent performance of the proposed MCFNN scheme.