A niching PSO algorithm based on clustering

For searching multi-maximum points of multi-modal functions, by analyzing standard particle swarm optimizer, a new niching method for particle swarm optimizer was proposed, which could identify and track global and local optima in a multi-modal search space. The sub-populations which represent the groups of particles specialized on niches were dynamically identified using density-based clustering algorithms. With this multi-population strategy, the diversity within the population was preserved and all the global/local optima were identified directly without further post-processing. Test solutions illustrate that the presented algorithm is efficient for both one-variable functions and multi-variable functions. 6figs., 10refs.