Passivity-Based Partial Sequential Model Predictive Control of T-type Grid-Connected Converters with Dynamic Damping Injection

LCL-filter interfaced three-level T-type converters have been widely used in low-voltage applications due to their elevated power quality. However, there exist several problems. To begin with, the performance of the finite-control-set model predictive control (FCS-MPC) controlled 3LT2C relies on an accurate system model and measurement. Consequently, output performances will be influenced when model parameter mismatches, grid impedance variation, and the dead-time of power devices occur. In addition, the robustness of the FCS-MPC controller may also encounter challenges under disturbances and noise. Furthermore, LCL-filter has the resonance problems in the grid currents, which may endanger the system stability. To solve these problems, a passivity-based partial sequential MPC (PSMPC) with dynamic-damping injection method, which outperforms FCS-MPC and ensures the asymptotic stability, is proposed. First, the passive output voltage based on the Euler-Lagrange model is obtained using dynamic damping injection. Second, by embedding passive output voltage into MPC, a passivity-based PSMPC robust controller is designed to enhance its anti-disturbance abilities and achieve resonance suppression. Finally, to avoid the time-consuming of weighting factor selection, a PSMPC that allocates nonconflicting objectives in the same layer is introduced. Experimental results on 10-kW prototype demonstrates its excellent performance over existing techniques.