Comparative Study of Optimization Techniques in Sizing Mesostructures for Use in NonPneumatic Tires

Abstract

This paper presents a comparative study on the size optimization of four mesostructures to meet the design requirements for a nonpneumatic tire's (NPT) shear beam with 10% shear flexure (SF) at 10 MPa effective shear modulus. The need for such comparison is motivated from the previous research wherein a systematic design method is proposed to design a mesostructure with high SF by studying the strain energy distribution pattern in honeycomb, auxetic, and sinusoidal auxetic mesostructure. Based upon the distribution pattern, a new type of mesostructure termed Stype is invented. Although it exhibited higher SF than the other mesostructures for a comparable set of geometry, it is not possible to validate the design method without exploring the complete design space of both existing and newly invented mesostructures. In order to address this limitation, these four mesostructures are optimized using the following optimization algorithms: (i) particle swarm; (ii) genetic algorithm (GA); and (iii) FAST–SIMPLEX (using response surface method). The results show the Stype mesostructure can be configured to meet the design requirements, thereby validating the design method presented in previous research. Additionally, it is also observed that auxetic mesostructure is only 5% less than the required design target, which presents an opportunity in future, to develop an alternate design method to maximize SF other than the one that is being validated in this paper.