文献类型：期刊文献

中文题名：基于核PLS方法的非线性过程在线监控

英文题名：Online nonlinear process monitoring using kernel partial least squares

作者：胡益[1];王丽[2];马贺贺[1];侍洪波[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;[2]上海应用技术学院电气与电子学院,上海200235

年份：2011

卷号：62

期号：9

起止页码：2555

中文期刊名：化工学报

外文期刊名：CIESC Journal

收录：CSTPCD;;EI(收录号：20113814343945);Scopus;北大核心:【北大核心2008】；CSCD:【CSCD2011_2012】；

基金：上海市重点学科建设项目(B504);国家自然科学基金项目(61074079)~~

语种：中文

中文关键词：核偏最小二乘;过程监控;非线性过程;质量预测

外文关键词：kernel partial least squares; process monitoring; nonlinear process; quality prediction

摘要：针对过程监控数据的非线性特点,提出了一种基于核偏最小二乘(KPLS)的监控方法。KPLS方法是将原始输入数据通过核函数映射到高维特征空间,然后在高维特征空间再进行偏最小二乘(PLS)运算。与线性PLS相比,KPLS方法能充分利用样本空间信息,建立起输入输出变量之间的非线性关系。与其他非线性PLS方法不同,KPLS方法只需要进行线性运算,从而避免非线性优化问题。在对过程进行监控时,首先采用KPLS方法建立模型,得到得分向量,然后计算出T2和SPE统计量及其相应的控制限。Tennessee Eastman(TE)模型上的仿真研究结果表明,所提方法比线性PLS方法具有更好的过程监控性能。
To handle the nonlinear problem for process monitoring,a new technique based on kernel partial least squares（KPLS）is developed.KPLS is an improved partial least squares（PLS）method,and its main idea is to first map the input space into a high-dimensional feature space via a nonlinear kernel function and then to use the standard PLS in that feature space.Compared to linear PLS,KPLS can make full use of the sample space information,and effectively capture the nonlinear relationship between input variables and output variables.Different from other nonlinear PLS,KPLS requires only linear algebra and does not involve any nonlinear optimization.For process data,firstly KPLS was used to derive regression model and got the score vectors,and then two statistics,T^2 and SPE,and corresponding control limits were calculated.A case study of the Tennessee-Eastman（TE）process illustrated that the proposed approach showed superior process monitoring performance compared to linear PLS.