PROJECT TITLE :

A General Framework of Multipopulation Methods With Clustering in Undetectable Dynamic Environments

ABSTRACT :

To solve dynamic optimization issues, multiple population ways are used to reinforce the population diversity for an algorithm with the aim of maintaining multiple populations in numerous subareas in the fitness landscape. Many experimental studies have shown that locating and tracking multiple relatively good optima instead of one world optimum is an efficient plan in dynamic environments. However, many challenges need to be addressed when multipopulation methods are applied, e.g., how to form multiple populations, how to keep up them in numerous subareas, and how to house matters where changes can't be detected or predicted. To handle these issues, this paper investigates a hierarchical clustering method to find and track multiple optima for dynamic optimization problems. To house undetectable dynamic environments, this paper applies the random immigrants technique while not amendment detection based mostly on a mechanism which will automatically reduce redundant individuals within the search house throughout the run. These ways are implemented into several research areas, as well as particle swarm optimization, genetic algorithm, and differential evolution. An experimental study is conducted primarily based on the moving peaks benchmark to test the performance with several alternative algorithms from the literature. The experimental results show the potency of the clustering technique for locating and tracking multiple optima as compared with other algorithms primarily based on multipopulation ways on the moving peaks benchmark.