摘要
针对模糊C-均值聚类算法存在对初始聚类中心敏感和聚类目标函数容易陷入局部最优的问题,提出了1种基于混沌差分进化模糊C-均值聚类的多模型建模方法。该方法采用混沌差分进化算法对模糊C-均值聚类的目标函数进行全局寻优,能有效的解决上述问题。将该方法应用于双酚A生产过程的质量指标软测量建模,仿真结果表明了该算法的有效性。
In view of the facts that the multiple models based on fuzzy C-mean clustering is sensitive to clustering center and the objection ftmction is inclined to fall into a local optimum, a novel fuzzy C-mean clustering based on Chaotic Differential Evolution, which is for multiple models soil-sensing modeling, is presented. The proposed algorithm optimizes objection function of fuzzy C-mean clustering by using Chaotic Differential Evolution and gets a global optimal solution, which can effectively address the problems mentioned above. Applying the algorithm to a soil-sensor model for the Bisphenol-A productive process, it is shown that the algorithm is effective.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2011年第3期291-294,共4页
Computers and Applied Chemistry
基金
国家自然科学基(60674092)
江苏省高技术研究项(BG20060010)
关键词
混沌差分进化
模糊C-均值聚类
多模型
软测量
chaotic differential, evolution, fuzzy C-mean clustering, multi-model, soft-sensor