3208 Robustness of Redundancy Optimization using Hierarchical Genetic Algorithms

概要

論文の詳細を見る
Most of the existing conventional optimization approach implicitly assumes that the system design process is deterministic and for a given input the optimized system will always produce the same output. However, in real world application, this is not true. In fact, in the real world, many engineering design optimization problems have parameters with uncontrollable variations due to noise or uncertainty. These variations can significantly degrade the performance of optimum solutions and can even change the feasibility of obtained solutions. Therefore, robust optimum solution is an important area of research in engineering design optimization. Many researchers in the past have investigated the impacts of uncertainties in design optimization techniques including evolutionary algorithms. However, hierarchical class of design optimization problems under uncertainty is not yet investigated using evolutionary techniques. In this direction, this paper attempted to explore the uncertainties in design parameters and its impacts on optimal solution using HGA. In order to achieve robust solution, the HGA has been applied to solve a hierarchical series redundancy optimization problem. The impacts of uncertainties on optimum solutions have been studied. In sum, the research work in this paper is an effort to develop a robust evolutionary algorithm for design optimization under uncertainties.
一般社団法人日本機械学会の論文
2006-11-14