A dual-layer intelligent adaptive optimization framework for optimizing the dynamics performance of automotive chassis

Vehicle dynamics performance critically influences driving comfort and safety, necessitating advanced optimization strategies. Artificial intelligence-based methods, despite their widespread use, face challenges such as poor surrogate model generalization, suboptimal algorithm convergence and limited robustness. To address these challenges, this article introduces a dual-layer intelligent adaptive optimization framework (DIAOF) to enhance automotive chassis dynamics. DIAOF integrates an intelligent adaptive ensemble model (IAEM) and a dual-layer multi-objective optimization algorithm (DMOA). IAEM uses intelligent model selection and linear regression to deliver precise, robust dynamics performance evaluation. DMOA uses a dual-layer mechanism and advanced population updates to balance global exploration and local exploitation, improving convergence and solution quality. Validated across various vehicle models, DIAOF achieves a 0.954 average prediction accuracy, a 5.81% performance improvement and reduced computational time compared to comparison methods, demonstrating superior generalization, efficiency and optimization performance.