A Serial Population Genetic Algorithm for Dynamic Optimization Problems
L. Zwanepol Klinkmeijer, Edwin D. de Jong, Wiering · 2006
The increase in number of papers being published on Evolutionary Algorithms for dynamic optimization problems shows the growing interest in this field. In this paper we introduce a powerful new genetic algorithm, the Serial Population Algorithm (SPA), which uses memory to solve dynamic tasks. We compare SPA to two other algorithms: a Hypermutation-and a Diploid Genetic Algorithm. Our results show that SPA outperforms these algorithms on a recurrent, discontinuous, dynamic optimization task. 1.