Structural optimization of bus chassis frame based on proxy model

Under the premise of ensuring modal and strength characteristics, achieving lightweight design of the structure simultaneously has become a key issue of concern for major automobile manufacturers and research institutions. To reduce the mass redundancy of the bus chassis frame, save production costs and energy consumption, a multi-objective optimization scheme based on surrogate model technology was proposed, which could maximize weight reduction without reducing the natural frequency or increasing the peak stress. According to the working principle, load characteristics and composition of the chassis frame, a parametric coupling model for modal and strength was constructed, and the stress, deformation, natural frequency and vibration mode characteristics of the overall structure were obtained. The dimensions of H-steel were determined as design variables, and the discrete mapping data sets of maximum stress, first-order natural frequency and mass were obtained through the Latin square design scheme. Parameters such as the coefficient of determination, adjusted coefficient of determination and root mean square error were selected as the standard evaluation indicators for the accuracy of the response surface model. The reliability of different surrogate models was compared and analyzed, and finally the Kriging model was adopted as the approximation function in the construction of the mathematical model. An optimized mathematical model was constructed to convert the modal and strength objectives into boundary conditions. The design variables meeting the optimization objectives were derived through the sequential quadratic programming algorithm. The results showed that, without reducing the requirements for strength and stiffness indicators, this optimization scheme could reduce the weight of the chassis frame by 9.94 %, which has good economic benefits and engineering value.