Artificial Glowworm Swarm Optimization(GSO) Algorithm is a new swarm intelligence algorithm. So far, the application of this algorithm is in the area of multiple signal source location, identification of odour sources, and hazardous spills and so on. But the history of the algorithm is short, there have many problems need to further research. For example, the algorithm existing slow convergence and low precision in optimizing the multimodal function, compare with other mature swarm intelligence algorithms, the application area of the algorithm is narrow. Aiming to these problems, the main work in the article is to improve the glowworm swarm optimization algorithm for optimizing the multimodal function, and use the improved glowworm swarm optimization algorithm to solve the combinatorial optimization problem and the clustering problem.The main results of this research are as follows:(1) Design a dynamically changing step strategy for glowworm swarm optimization algorithm, which provides the algorithm with effective dynamic adaptability. Experiments show that, the improved algorithm can effectively improve the glowworm swarm optimization algorithm to optimize the multi-modal function existing slow convergence and low precision problems.(2) Propose a discrete glowworm swarm optimization(DGSO) algorithm for solving the traveling salesman problem. Experimental results indicate that the proposed algorithm have good robustness, and is very competitive with other algorithms.(3) Propose a new approach for clustering analysis based on the Glowworm Swarm Optimization algorithm. The new algorithm are tested on three data sets, the experimental results show that the new algorithm has higher clustering result. |