文献类型：期刊文献

中文题名：叠加支持向量机及其在醋酸精馏软测量中的应用

英文题名：A Simple Modified Multi-kernel SVR and Its Application in Soft Sensor Computing of An Industrial Acetic Acid Distillation System

作者：李静[1];刘爱伦[1]

机构：[1]华东理工大学化工过程先进控制和优化技术教育部重点实验室,上海200237

年份：2013

卷号：39

期号：2

起止页码：200

中文期刊名：华东理工大学学报（自然科学版）

外文期刊名：Journal of East China University of Science and Technology

收录：CSTPCD;;Scopus;北大核心:【北大核心2011】；CSCD:【CSCD2013_2014】；

语种：中文

中文关键词：叠加模型;支持向量机;醋酸精馏;软测量

外文关键词：additive models; support vector regression; acetic acid distillation; soft sensor

摘要：针对在高维输入空间数据点的异常稀疏性(维数灾难)会导致支持向量机回归模型产生偏差的问题,提出了一种基于叠加模型的支持向量机回归方法———叠加支持向量机回归(AddS-VR)。AddSVR的实现是通过对每一维输入进行核化,然后将每一个核空间进行叠加得到,基于叠加模型可以克服维数灾难的问题,使得其在处理高维问题时估计偏差减小。为了更方便、迅速地实现AddSVR,还提出了对支持向量机的一种简化的二次规划描述。将AddSVR用于醋酸共沸精馏中塔底醋酸组分的预测,仿真实验结果表明,AddSVR模型与传统的SVR和最小二乘支持向量机回归(LS-SVM)模型相比有更好的预测效果。
Aiming at the problem that the sparsity of data in the high dimensional input space wilt lead lo the biased estimation of support vector regression （SVR）, an additive SVR is proposed in this paper. This algorithm is realized via the addition of the separated input spaces after kernelization. Thus, the curse of dimensionality can be overcome by the additive model such that the bias can be reduced for high dimensional regression problem. Moreover, a simplified quadratic programing （QP） formulation of SVR is proposed for easily constructing the additive SVR model. Finally, the proposed method is employed to predict the concentration of HAC in the bottom outlet. It is shown from the experimental results that the additive SVR model can attain better predication performance than traditional SVR and least quare support vector regression （LS-SVR）.