Composite Sliding Mode Control with Disturbance Rejection for M3C in Mining Area Wind Power

To address the degradation of steady-state performance and the limitations of conventional control strategies for modular multilevel matrix converters (M3Cs) in mining wind power applications under complex operating conditions, such as internal multi-frequency coupling constraints and power step changes, this paper proposes a super-twisting sliding mode control strategy for the M3C based on a nonlinear disturbance observer (NDO). First, a decoupled mathematical model of the M3C is established. A nonlinear disturbance observer is then designed to provide real-time online estimation of unmodeled system dynamics and external lumped disturbances. Subsequently, a super-twisting sliding mode controller based on an integral sliding surface is constructed, and the lumped disturbance estimated by the NDO is introduced into the control law as a feedforward compensation term. Using Lyapunov theory, the finite-time reachability of the super-twisting sliding variable is established, and the exponential asymptotic stability of the current-tracking error on the sliding surface is demonstrated. Finally, the effectiveness of the proposed control strategy is verified through simulations, ablation comparisons, and experiments. The ablation results further show that the NDO mainly improves disturbance estimation and compensation as well as transient recovery, whereas the STSMC mainly reduces control chattering and improves current quality, demonstrating their complementary and synergistic effects. The results indicate that the proposed control strategy provides good steady-state performance and improves the dynamic performance and robustness of the mining wind power M3C system under power fluctuations compared with conventional control methods.