The main spindle torque and tower thrust of the wind turbine are the key factors affecting the operation efficiency and structural safety of the wind turbine

and the nonlinear relationship between wind speed and power output makes the aerodynamic load prediction of the wind turbine more complicated.Based on the wind speed and power reference values

it constructs a prediction model of wind turbine spindle torque and tower thrust

and combines non-dominated sorting genetic algorithm

weighted sum method and Pareto front construction method to optimize the analysis.By comparing the three optimization methods

it evaluates the effect of wind turbine performance prediction.The results show that the prediction model can accurately estimate the spindle torque and tower thrust

and the NSGA-Ⅱ method is particularly prominent in multi-objective optimization

and the optimized prediction accuracy is significantly improved.