Multiobjective Model Predictive Control of LCL Grid-Connected Inverter Based on an Improved Gray Wolf Algorithm

To address the challenges of high-current harmonic distortion, excessive switching losses, and less desirable dynamic response in the single-phase LCL grid-connected inverter, this article proposes a multiobjective model predictive control (MOMPC) method based on an improved gray wolf optimization (IGWO) algorithm. The proposed method first establishes the mathematical model of the LCL inverter and formulates the cost function of FCS-MPC. Subsequently, the voltage of the capacitor is introduced to impose constraints on the cost function, thereby constructing the MOMPC to further enhance the control performance. Finally, the IGWO algorithm is employed to perform offline multiobjective optimization on current error, total harmonic distortion, and switching frequency to design the optimal weighting coefficients of MOMPC. In contrast to traditional MPC methods, the IGWO-based MOMPC can reduce total harmonic distortion, switching frequency, and control errors while enhancing system robustness and dynamic performance. The simulation and experimental results validate the feasibility and effectiveness of the proposed method.