Abstract: A multi-objective optimization method for bridge wind barrier parameters is proposed by combining the CFD simulation of vehicle-bridge-wind barrier system and the adaptive surrogate model. Firstly, a two-dimensional CFD model of the vehicle-bridge-wind barrier system was established to obtain the aerodynamic coefficients of the train and bridge under different wind barrier parameters, and the calculation accuracy of the numerical model was verified by comparing with the wind tunnel test results. Secondly, the wind barrier height and air permeability were selected as design parameters and the candidate sample point set was obtained. The optimal Latin Hypercube experimental design method was used to select the initial sample point set of optimization parameters. By taking the minimization of bridge resistance coefficient and train overturning moment coefficient as the optimization objectives, the objective function surrogate model is quickly constructed by adaptive sampling technology. Finally, based on the multi-objective Grey Wolf optimization algorithm, the Pareto optimal solution sets of the two objective functions are calculated. The individuals in Pareto solution set are scored and ranked by entropy weight TOPSIS comprehensive evaluation method, from which the optimal solution is selected and the corresponding wind barrier parameters are obtained. The results show that the wind barrier height has a great influence on the aerodynamic force of the bridge, and the wind barrier permeability has a greater influence on the aerodynamic force of the train. The wind barrier parameter with the height of 3.37 m and the ventilation rate of 0% has the best wind resistance effect on the bridge and the train. This study can provide reference for the optimization of wind barrier parameters of various bridges and related research.