In this work, a multiobjective aerodynamic optimization of a high-speed train head was performed to improve the aerodynamic performance of the high-speed train running on an embankment under crosswinds. Seven optimization design variables were selected to control five regions on the train head. The total aerodynamic drag force, aerodynamic lift force, and aerodynamic side force of the head coach were set as the optimization objectives. The optimal Latin hypercube sampling method was used to obtain the values of the design variables of the sample points. The high-speed train head was deformed using the free-form deformation approach through which the mesh morphing was performed without remodeling and re-meshing. Then, the aerodynamic performances of the high-speed trains at the sample points were calculated using the computational fluid dynamics method. A Kriging surrogate model between the design variables and their optimization objectives was constructed. Then, the multiobjective aerodynamic optimization of the high-speed train head was performed using multiobjective genetic algorithms based on the Kriging model. Based on the results of the sample points, the relationships between the optimization design variables and the optimization objectives were analyzed, and the contributions of the primary factors to the optimization objectives were obtained. After optimization, a series of Pareto-optimal head shapes were obtained. The steady and unsteady aerodynamic performances of the train with an optimal head, which was selected from the Pareto-optimal head shapes, were compared with those of the original train.
An efficient anomaly detection for high-speed train braking system using broad learning system
