Improvement of Computing Efficiency and Regularization of Inner Boundary Shape Optimization in Temperature Field.

Abstract

The purpose of this study is the improvement of the efficiency of design process that reaches to detail design from initial design. Genetic algorithm is applied to the initial design stage and conjugate gradient method is applied to the detail design stage. The design solution obtained by GA can be adopted as an initial design solution in the detail design stage because GA has the advantage of the capability of the global search in the optimization process. Furthermore, the regularization method is applied as an avoidance strategy of the ill-posedness that encounters at the process of the detail shape determination. The effectiveness of the proposed method is examined through an application to the optimization problem of the inner boundary shape determination in temperature field.