Frequency and severity are a priori very influential parameters in the performance of Dynamic Optimization Problems because they establish when and how hard is the change of the target optimized function. We study in a systematic way their influence in the performance of Dynamic Optimization Problems and the possible mathematical correlations between them. Specifically, we have used a steady state Genetic Algorithm, which has been applied to three classic Dynamic Optimization Problems considering a wide range of frequency and severity values. The results show that the severity is the more important parameter influencing the accuracy of the algorithm.
展开▼
