A novel lightweight combinatorial optimization strategy based on ASA-NLPQL optimization algorithm for front seat skeleton of a passenger car

Abstract

This study presents a novel combinatorial optimization strategy to address the challenges of lightweight design for automotive seat skeletons. The proposed approach integrates surrogate modeling with a hybrid optimization algorithm to enhance computational efficiency and solution accuracy. The hybrid algorithm leverages the strengths of global and gradient-based optimization methods by combining Adaptive Simulated Annealing and Non-Linear Programming by Quadratic Lagrangian (ASA-NLPQL). The effectiveness of the ASA-NLPQL hybrid algorithm is validated through two numerical function examples and an engineering optimization case study, and it is subsequently applied to a real-world case study focused on the lightweight optimization design of an automotive seat skeleton. Firstly, the finite element model for modal analysis of the entire front seat skeleton of a passenger car is developed and validated through experimental tests. Key design variables are identified via sensitivity analysis to guide the optimization process. Subsequently, a surrogate model is constructed using sample points generated by optimal Latin hypercube sampling. The combined optimization strategy, based on the surrogate model and hybrid algorithm, is then applied to the optimal design of the passenger car seat skeleton. Optimization results demonstrate a 22.3% reduction in seat weight with negligible impact on vibration performance. Finally, the optimized seat skeleton undergoes various strength tests, with the results confirming that it meets the required strength criteria. These findings demonstrate the effectiveness of the proposed optimization strategy and provide a reliable reference for similar engineering optimization challenges.