Improved whale optimization algorithm and its application in vehicle structural crashworthiness

Abstract

Whale optimization algorithm (WOA) is a novel-innovative swarm-based meta-heuristic algorithm with excellent performance, but it may still be trapped into local extremum for troublesome problems. To this end, an improved multi-objective whale optimization algorithm (IMOWOA) is proposed to cover the shortages. Firstly, in the search stage of WOA, individual difference is considered to strengthen the exploration ability, and evolution operators are introduced to regenerate the stagnated population to prevent premature convergence. Next, the performance of IMOWOA is compared with MOWOA and other classical optimization algorithms, and a series of multi-objective test functions are used. The results on the convergence and diversity of Pareto front confirm that IMOWOA has better feasibility and competitiveness. Finally, integrated with the least squares support vector regression (LSSVR) model, IMOWOA is applied to the deterministic optimization of vehicle structural crashworthiness. The conclusion testified the efficiency of IMOWOA in the field of vehicle crashworthiness.

Keywords:

• Improved multi-objective whale optimization
• individual difference
• evolution operators
• vehicle structural crashworthiness

Data availability

The source codes of WOA are available at http://www.alimirjalili.com/WOA.html. The source codes of MOPSO, MOEA/D and NSGA-II are available at https://emoo.cs.cinvestav.mx/. The Additional data that support the findings of this study is available from the corresponding author upon reasonable request.

Additional information

Funding

This work is supported by Anhui Provincial Development and Reform Commission new energy vehicle and intelligent network connected automobile industry technology innovation project.