Shape Optimization of the Streamlined Train Head for Reducing Aerodynamic Resistance and Noise

Aiming to improve the comprehensive aerodynamic performance of a high-speed train, multi-objective shape optimization method for streamlined train head is proposed in this work. The parameterized with some spline curves. design variables are uniformly sampled using optimal Latin hypercube method. resistance and dipole noise sources chosen as objectives, which can be obtained through computational fluid dynamics (CFD) An approximate calculation model established by radial basis function neural network so effectively predict values objectives. error between predicted actual less than 1%, that source 3 dB, demonstrate validity model. In process, algorithm NSGA-II adopted update variables, used calculate greatly reduces computation time shape. Through iterative space, each optimized variable shows trend convergence, generally show decreasing trend. Pareto front corrected CFD after optimization. reduced up 4.5%, 3.9 dB.