Research on crashworthiness and lightweight of B-pillar based on MPSO with TOPSIS method

Abstract

To better improve the lightweight and crashworthiness performance of B-pillar, this paper presents a hybrid approach combining the modified particle swarm optimization algorithm (MPSO) and the technique for order preference by similarity to ideal solution (TOPSIS) approach with principal component analysis method (PCA) has been applied to multi-objective lightweight optimization design of the outer panel of the B-pillar structure of automotive. First, the numerical model was conducted to investigate the crashworthiness indicators of H-point under the side collision situation, which agree well with experimental results. Second, a steel–aluminum B-pillar was constructed and through integrating experimental design method with a surrogate model based on the numerical model to construct the approximate model, MPSO algorithm was adopted for multi-objective optimization design of B-pillar and the Pareto solutions set was determined. Finally, TOPSIS with PCA method can obtain the best compromise solution. It is observed that the weight of the B-pillar is reduced by 20.7% under the precondition of the peak of impacting velocity and the peak intrusion deformation has improved by 23.5% and 41.7%, respectively. Moreover, a comparison of the optimal design and original model, it reveals that the proposed hybrid method has a superior performance used for multi-objective lightweight optimization design of B-pillar structure.