Data-Driven Surrogate Model for Automobile Wheel Strength and Stiffness Based on Deep Learning

Structural strength and stiffness are critical factors in the design of automotive wheel spokes. In the context of conceptual design innovation, developing data-driven surrogate models for performance simulation is essential to meet requirements of rapid iterations. To address the issue of low accuracy caused by poorly distributed design samples, this paper proposes a method that combines the generation of high-quality samples with a deep regression model for surrogate modelling. Comparative experiments demonstrate that the proposed method outperforms its alternatives in terms of accuracy and efficiency, effectively providing support for innovative wheel spoke design.