A New Advantage Sharing Inspired Particle Swarm Optimization Algorithm

摘要

Particle swarm optimization algorithm is a widely used computational method for optimizing a problem. This algorithm has been applied to many applications due to its easy implementation and few particles required. However, there is a big problem with the PSO algorithm, all the virtual particles converged to a point which may or may not be the optimum. In the paper, we propose an improved version of PSO by introducing the idea of advantage sharing and pre-learning walk mode. The advantage sharing means that the good particles share their advantage attributes to the evolving ones. The pre-learning walk mode notices one particle if it should continue to move or not which uses the feedback of the last movement. Two more algorithms are simulated as the comparison methods to test Benchmark function. The experimental results show that our proposed scheme can converge to a better optimum than the comparison algorithms.