Open AccessThe study of parametric optimization algorithms on example of vehicle bumper crashworthiness
Author(s) -
Roman Goncharov,
Valeri Zuzov
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/820/1/012030
Subject(s) - crashworthiness , radial basis function , parametric statistics , polynomial , metamodeling , mathematical optimization , optimization problem , kriging , algorithm , reduction (mathematics) , polynomial regression , computer science , nonlinear system , engineering , mathematics , regression analysis , artificial neural network , finite element method , structural engineering , artificial intelligence , machine learning , mathematical analysis , statistics , physics , geometry , quantum mechanics , programming language
Different algorithms of parametric optimization (implemented in the program LS-OPT) to solve high-speed and high-nonlinear problems of impact character on the example of vehicle bumper optimization are considered and compared. Different algorithms of response surface optimization such as linear polynomial model, quadratic polynomial model, Feedforward, Radial Basis Function, Kriging and Support Vector Regression were studied and analyzed; algorithms of choosing point selection scheme were considered. The analysis of optimization results allowed us to determine which of these algorithms are most effective in terms of accuracy and computational time. The problem of impact on the vehicle’s bumper is solved in the paper, the optimization is based on the application of metamodel (RBF). The method provided a reduction of its mass by 16% while maintaining the initial parameters of crashworthiness.