Adaptive-Gain Second-Order Sliding-Mode Control of NPC Converters Via Super-Twisting Technique

In this article, a two-stage design approach combining an adaptive-gain generalized super-twisting algorithm (GSTA) and a modified super-twisting observer (STO) is proposed for the three-level neutral-point-clamped (NPC) converters. The key-point of this method is to design some dynamically adapted gains in the standard GSTA, and the tradeoff between dynamic behavior and chattering phenomenon can be addressed. Consequently, the dynamic and steady-state performance of the NPC converter is improved by applying the adaptive-gain GSTA in the power tracking loop and the voltage regulation loop simultaneously. On the other hand, since the convergence rate of the conventional STO is limited while the system trajectories are far away from the origin, a modified STO is applied in the voltage regulator of the NPC converters to reject external disturbances. The finite-time stability of the proposed control scheme is proved. Finally, several comparative experiments are carried out based on a real three-level NPC converter, and the feasibility and effectiveness of the proposed method are validated.