上海交通大学学报

上海交通大学学报（自然版） ›› 2011, Vol. 45 ›› Issue (08): 1172-1175.

• 自动化技术、计算机技术 • 上一篇    下一篇

一种带监督的仿射传播聚类多模型建模方法

邓卫卫1，杨慧中1,2   

  1. （1.江南大学 教育部轻工过程先进控制重点实验室，江苏 无锡 214122；2.上海市电站自动化技术重点实验室，上海 200072）
  • 收稿日期:2011-03-22 出版日期:2011-08-30 发布日期:2011-08-30
  • 基金资助:

    国家自然科学基金资助项目(60674092),江苏省高技术研究项目(BG2006010),上海市科学技术委员会资助项目(09DZ2273400)

A Multi-model Modeling Method Based on Supervised Affinity Propagation Clustering Algorithm

 DENG  Wei-Wei-1, YANG  Hui-Zhong-1, 2   

  1.  (1. Key Laboratory of Advanced Process Control for Light Industry (Ministry of Education), Jiangnan University, Wuxi 214122, China; 2. Shanghai Key Laboratory of Power Station Automation Technology, Shanghai 200072, China)
  • Received:2011-03-22 Online:2011-08-30 Published:2011-08-30

摘要：  针对传统的多模型建模方法在聚类过程中不考虑模型的输出误差而导致最终的模型存在较大误差的问题，提出了一种带监督的仿射传播聚类多模型建模方法.该方法先由仿射传播聚类算法得到初始聚类，然后，根据输出误差对聚类进行循环调整至各类别不再变化为止，最后，得到准确划分的聚类并采用最小二乘支持向量机建立子模型来实现对输出的估计，并将本文的建模方法应用到某双酚A反应釜出口丙酮含量的软测量建模中进行仿真.结果表明，该方法可以获得比传统的多模型建模方法更好的建模效果.

关键词: 多模型, 监督, 仿射传播聚类, 最小二乘支持向量机

Abstract: The traditional multimodel modeling method has big error because it does not consider the output error of the model during the clustering process, A multi-model modeling method based on supervised affinity propagation clustering algorithm was proposed. The principle is that the initial clusters are first obtained by the affinity propagation clustering algorithm, and then the clusters are adjusted cycledly in accordance with the output errors until the clusters do not change. Finally, the accurate clusters are got, and the sub-models are respectively built by least squares support vector machine so as to estimate the output. The method is used for the softsensor model to estimate the content of acetone at the outlet of a reaction vessel in a bisphenol A production process. The simulation results show that the method can get better modeling results than the traditional multi-model modeling method.

Key words:  multi-models, supervised, affinity propagation clustering, least squares support vector machine (LSSVM)

中图分类号: