GENETIC ALGORITHM OPTIMIZATION OF MULTI-PEAK PROBLEMS - STUDIES IN CONVERGENCE AND ROBUSTNESS

Engineering design studies can often be cast in terms of optimization problems. However, for such an approach to be worthwhile, designers mu st be content that the optimization techniques employed are fast, accu rate and robust. This paper describes recent studies of convergence an d robustness problems found when applying genetic algorithms (GAs) to the constrained, multi-peak optimization problems often found in desig n. It poses a two-dimensional test problem which exhibits a number of features designed to cause difficulties with standard GAs and other op timizers. The application of the GA to this problem is then posed as a further, essentially recursive problem, where the control parameters of the GA must be chosen to give good performance on the test problem over a number-of optimization attempts. This overarching problem is de alt with both by the GA and also by the technique of simulated anneali ng. It is shown that, with the appropriate choice of control parameter s, sophisticated niche forming techniques can significantly improve th e speed and performance of the GA for the original problem when combin ed with the simple rejection strategy commonly employed for handling c onstraints. More importantly, however, it also shows that more sophist icated multi-pass, constraint penalty functions, culled from the liter ature of classical optimization theory, can render such methods redund ant, yielding good performance with traditional GA methods.