Cultural-Based Particle Swarm Optimization for Dynamical Environment

Abstract

In this paper, we have presented a new method of the Cultural-based Particle Swarm Optimization for Dynamical Environment which uses belief space representation to accommodate more types of knowledge such as History Knowledge, and Topographical Knowledge. Also, additional influence functions have been developed to utilize these types of knowledge. This study focuses on the knowledge needed to track an optimal solution in an environment where the functional landscape is produced by the combination of n overlapped cones in a 2 dimensional grid. Then, the parabolic benchmark functions with various severities are used to test, compared with the PSO, and the results show the modified strategies are effective in tracking changes.