Aerodynamic modal analysis of racing car using multiobjective genetic algorithm

In recent years, surrogate model techniques have found extensive applications in engineering fields such as automotive traffic, building energy, and communication networks. Therefore, surrogate model techniques have become particularly important in addressing complex real-world engineering problems. This study employs a Kriging surrogate model for response surface fitting. An optimization mathematical model of the chassis is established, and the NSGA-Ⅱ multiobjective genetic algorithm is selected for chassis optimization design. After multiple iterations, a set of optimal solutions is obtained, which are then rounded for performance validation and comparison with the original chassis. The results show that the chassis achieved a weight reduction of 10.63%. Although the chassis displacement and stress increased in some conditions, they remained within acceptable safety limits. Furthermore, the torsional stiffness of the chassis was increased by 44.05%. Collision analysis of the lightweighted chassis indicates high safety performance. Additionally, the high-speed obstacle avoidance test results for the actual vehicle were improved by 3 seconds, and the 8-loop test results were improved by 0.3 seconds. Due to time constraints, this paper only considers the frontal 100% collision when carrying out the collision analysis of the lightweight frame, and the subsequent frontal offset collision and side collision can be carried out. The collision analysis under multiple working conditions can ensure the safety performance of the frame.