Estimation of Rubber Material Property by Successive Zooming Genetic Algorithm

Abstract

The industrial use of various kinds of rubber-like (hyper-elastic) material is rapidly increasing and growing in importance, especially in automobiles, trains, and machinery(1). In the past, rubber engineers and designers have predicted the behavior of rubber-like materials using analytic methods for limited problems or approximate methods for general problems. Yet, with the progress of digital computers, finite element methods(2), represented by the Mooney-Rivlin model, are now widely used to analyze hyper-elastic as well as isotropic materials.The conventional method used to evaluate the properties of rubber-like materials is the least square method (LSM), however, this method has a low precision and involves a tedious pre-solving process. Accordingly, this study proposes a simple yet powerful method for estimating the properties of rubber-like materials using a successive zooming genetic algorithm (SZGA). The proposed method results in dependable and precise rubber-like properties for various Mooney-Rivlin models based on simply changing the objective function. To demonstrate the effectiveness of the proposed method, it is compared with Haines & Wilson's method (LSM) and other commercial packages.