Clustering research of fabric deformation comfort using bi-swarm PSO algorithm

This paper proposes a new approach to using particle swarm optimization (PSO) algorithm to cluster fabric comfort. It is shown how PSO can be used to cluster the deformation comfort data by according to the similarity of the deformation comfort characteristics and the desired cluster number. The swarm was divided into two subgroups, and the inertia weight of each subgroup dynamically changed along with the iterative generations and fitness value respectively. The new algorithm was evaluated on data sample, and the clustering center was seen as the solution of the particle. The analysis of the clustering results and the comparison of fuzzy cluster results and PSO-based cluster results show that the proposed algorithm has great practical value and ability to overcome the disadvantages of fuzzy cluster which depends on the human experience and cannot cluster according to the desired cluster number directly.

[1]  R. Eberhart,et al.  Empirical study of particle swarm optimization , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[2]  Andries Petrus Engelbrecht,et al.  Data clustering using particle swarm optimization , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[3]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[4]  Russell C. Eberhart,et al.  A new optimizer using particle swarm theory , 1995, MHS'95. Proceedings of the Sixth International Symposium on Micro Machine and Human Science.

[5]  Thomas E. Potok,et al.  Document clustering using particle swarm optimization , 2005, Proceedings 2005 IEEE Swarm Intelligence Symposium, 2005. SIS 2005..

[6]  Chen Pei-gen.  Application of fuzzy cluster analysis in fabric variation comfortability evaluation , 2006 .

[7]  R. Eberhart,et al.  Comparing inertia weights and constriction factors in particle swarm optimization , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).