Abstract

This study proposes a new kind of optimization that combines particle swarm optimization with a filter approach called FM_PSO. To enhance the first stages exploitation potential FM_PSO uses the filter approach which may expand the iteration acceptance criteria. At the same time the split-and- conquer strategy is being examined as a way to accelerate the pace of convergence. FM PSO may enhance exploration and exploitation capabilities even after optimizing on a massive scale. Based on the results of the numerical testing FM_PSO is the best method when it comes to the macromolecular structure prediction model and the test function.