×

A PSO algorithm with high speed convergence. (Chinese. English summary) Zbl 1224.68156

Summary: For the problem that particle swarm optimization (PSO) algorithms often suffer from being trapped in local optima (premature convergence), an improved PSO with high-speed convergence is proposed to efficiently control premature stagnation. Firstly, chaotic sequence is used to initiate individual position, which strengthens the diversity of searching. Furthermore, an effective method that identifies premature stagnation is embedded into PSO, so once premature stagnation happens, a randomized solution, as a substitute for current optimum, is used to change the current searching locus so that particles can go out of the local optima. By using the two measures, the searching process can converge to the global optimum with high speed. Abundant simulation experiments demonstrate that the algorithm proposed in this paper only needs several particles and iterates a few times to be able to obtain the global optimum for most continuous function optimization problems. The convergence speed and searching ability are quite outstanding and satisfactory.

MSC:

68T20 Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.)
PDFBibTeX XMLCite