Abstract
Purpose - This paper aims to describe a shape optimization for hyperelastic axisymmetric structure with an exact sensitivity method.
Design/methodology/approach - The whole shape optimization process is carried out by integrating a closed geometric shape in the real space R-2 with boundaries defined by B-splines curves. An exact sensitivity analysis and a mathematical programming method (SQP: Sequential Quadratic Programming) are implemented. The design variables are the control points' coordinates which minimize the Von-Mises criteria, with a constraint that the total material volume of the structure remains constant. The feasibility of the proposed methods is carried out by two numerical examples. Results show that the exact Jacobian has an important computing time reduction.
Findings - Numerical examples are presented to illustrate its performance.
Originality/value - In this work, the sensitivity performance is computed using two numerical methods: the efficient finite difference scheme and the exact Jacobian.