Research on clustering method of resource service cluster in cloud manufacturing

In view of the fuzzy classification boundaries of massive and diverse cloud manufacturing (CMfg) service resources, this paper analyzes the relationship between cloud services and manufacturing resources, and establishes a CMfg hybrid service aggregation model based on the service resources aggregation type. In addition, this paper establishes a clustering validity evaluation function based on k-means clustering algorithm. Aiming at the disadvantage that k-means clustering algorithm is sensitive to the initial clustering center, the shuffled frog leaping algorithm (SFLA) is introduced to determine the initial clustering center. The inverse solution is used to expand the search range of the initial frog population, and the optimization of the worst frog population is improved by combining the mean value of the optimal solution. Based on the improved leapfrog algorithm and k-means iteration, an improved k-means clustering algorithm based on leapfrog algorithm is proposed. Finally, the validity of the algorithm is verified by the Iris test data set and a self-constructed data set (Self-cd), and the feasibility of the algorithm is illustrated by the application of lathe resource on the CMfg platform.