设为首页 联系我们 加入收藏

当前位置: 网站首页 期刊分类目录 2012第5期 正文

基于自适应仿射传播聚类算法的多模型建模方法

作者:时间:2014-06-19点击数:

 全文下载:2012050515

郝美玉,田学民*,王平

 (中国石油大学(华东) 信息与控制工程学院,山东 青岛 266580)

 摘要:考虑到工业生产数据具有按工作点聚类和迁移的特点,提出了一种基于自适应仿射传播聚类 (adAP)的多最小二乘支持向量机(LSSVM)算法进行软测量建模。该方法用adAP算法对训练样本进行分类以找到最优的聚类结果,采用LSSVM算法对各类样本分别建立子模型,并根据当前工作点所属子类的模型进行预测输出。将该方法用于聚丙烯熔融指数的软测量建模,结果表明,与其他方法相比该方法具有更高的回归精度和良好的泛化能力。

 关键词:软测量; 多模型; 自适应仿射传播聚类算法; 最小二乘支持向量机

 中图分类号: TP 391文献标志码: A

 Multiple Models Soft-sensing Method Based on Adapt Affinity Propagation

 HAO Mei-yu, TIAN Xue-min, WANG Ping

 College of Information and Control Engineering, China University of Petroleum, Qingdao 266580,China)

 Abstract: Since the industrial production samples are clustered around different operating points, a soft-sensing method with multiple models based on Adaptive Affinity Propagation Clustering Algorithm (adAP) and Least Square Support Vector Machine (LSSVM) is proposed. Classify the training samples into several classes using the adAP clustering to find the best clustering result, and train the sub-models by LSSVM according to corresponding sub-class samples. The test samples are assigned to appropriate sub-class, then predicted outputs are estimated by corresponding sub-models. The simulation results of Melt Index indicate that the proposed method has better prediction accuracy and generalization performance.

 Key words: soft-sensing, multiple model, adaptive affinity propagation clustering algorithm, least square support vector machine

 收稿日期:2012-07-01

 基金项目: 国家自然科学基金项目(51104175),山东省自然科学基金项目(ZR2011FM014)

 作者简介: 郝美玉(1989—),女,硕士研究生. *通信联系人.

Copyright © 2011-2017 青岛科技大学学报 (自然科学版)