Computational Mechanics using High Performance Computing
Chapter 13
G. Thierauf and J. CaiDepartment of Civil Engineering, University of Essen, Essen, Germany By introducing a variable coding technique, a parallel optimization method based on a combination of GAs and ESs is presented. The advantages of both GAs and ESs, such as coding of genetic information and adaptation of optimization parameters, are enhanced by this new method. Enhanced by the availability of high-speed parallel computing technique, zero-order and direct optimization methods gained renewed interest since the early eighties. Influenced by this development, stochastic search methods and Darwinian methods have been the subject of many publications during the past fifteen years. Among these, the genetic algorithms (GAs) and the evolution strategies (ESs) are two of the most discussed search strategies. In this Chapter, the basic concepts of GAs and ESs are first briefly described. Subsequently, by introducing a variable coding technique, a parallel optimization method based on a combination of GAs and ESs is presented. | ||||||||||
