Abstract: Addressing the issues of power fluctuations and abrupt load changes in wind turbines during low-voltage ride-through (LVRT), this study proposes an adaptive optimization strategy based on model predictive control (MPC) for the LVRT process. First, a theoretical coupled load model of the turbine unit under LVRT conditions is established to elucidate the torsional vibration characteristics of the drive train and the transmission mechanism of aerodynamic loads during grid voltage dips. Building on this foundation, an MPC-based adaptive optimization strategy is introduced to mitigate power oscillations and sudden load variations during LVRT. A co-simulation platform integrating MATLAB/Simulink electromechanical transient models and Bladed coupled models is developed to comparatively analyze the overall operational characteristics and load behaviors under traditional control strategies versus the proposed MPC-based optimization strategy. Experimental results demonstrate that under 0.2 p.u. LVRT transient conditions, the proposed strategy reduces turbine power fluctuations by 15% and decreases extreme loads on critical components by 8%. This effectively validates the technical advantages of the control framework in enhancing the fault ride-through robustness of turbine units and improving active grid voltage support capabilities.