Dynamic hypersphere based support vector data description for batch process monitoring

Abstract Support Vector Data Description (SVDD) is an efficient monitoring method that captures the spherically shaped boundary around the normal batch data and sets the control limit related to support vectors (SVs) for online monitoring. Using nonlinear transformation functions, SVDD constructs an irregular hypersphere in high dimensional space. When the batch process is complicated, the accuracy of monitoring will decrease with traditional control limit of SVDD. In this paper, dynamic hypersphere based support vector data description (DH-SVDD) is proposed for batch process monitoring. In training process, static hypersphere is built by the important SVs of training dataset. In testing process, dynamic hypersphere is built by the important SVs of combined dataset with current test sample and training dataset. If there is a significant change between these two hyperspheres, it means that the current test sample is an outlier. Thus, DH-SVDD has a relatively high monitoring accuracy because it fully considers relationship between the current test sample and the historical training dataset in high dimensional space. Comparison is conducted between the proposed DH-SVDD and traditional methods such as K-chart-SVDD, max limit SVDD and validation limit SVDD. The effectiveness of the DH-SVDD is also verified by a semiconductor etch process and a fed-batch penicillin fermentation process.

[1]  Svante Wold,et al.  Modelling and diagnostics of batch processes and analogous kinetic experiments , 1998 .

[2]  Barry M. Wise,et al.  A comparison of principal component analysis, multiway principal component analysis, trilinear decomposition and parallel factor analysis for fault detection in a semiconductor etch process , 1999 .

[3]  Xuefeng Yan,et al.  Independent component analysis-based non-Gaussian process monitoring with preselecting optimal components and support vector data description , 2014 .

[4]  Seoung Bum Kim,et al.  One-class classification-based control charts for multivariate process monitoring , 2009 .

[5]  Xuefeng Yan,et al.  Just‐in‐time reorganized PCA integrated with SVDD for chemical process monitoring , 2014 .

[6]  Seoung Bum Kim,et al.  A Density‐focused Support Vector Data Description Method , 2014, Qual. Reliab. Eng. Int..

[7]  Theodora Kourti,et al.  Multivariate dynamic data modeling for analysis and statistical process control of batch processes, start‐ups and grade transitions , 2003 .

[8]  Jie Yu.  Multiway Gaussian Mixture Model Based Adaptive Kernel Partial Least Squares Regression Method for Soft Sensor Estimation and Reliable Quality Prediction of Nonlinear Multiphase Batch Processes , 2012 .

[9]  Gülnur Birol,et al.  A modular simulation package for fed-batch fermentation: penicillin production , 2002 .

[10]  Zhen Zhao,et al.  Multi-phase MPCA modeling and application based on an improved phase separation method , 2012 .

[11]  Si-Zhao Joe Qin,et al.  Survey on data-driven industrial process monitoring and diagnosis , 2012, Annu. Rev. Control..

[12]  Bhavik R. Bakshi,et al.  Analysis of operating data for evaluation, diagnosis and control of batch operations , 1994 .

[13]  Robert P. W. Duin,et al.  Support Vector Data Description , 2004, Machine Learning.

[14]  J. Macgregor,et al.  Monitoring batch processes using multiway principal component analysis , 1994 .

[15]  ChangKyoo Yoo,et al.  On-line monitoring of batch processes using multiway independent component analysis , 2004 .

[16]  Furong Gao,et al.  Review of Recent Research on Data-Based Process Monitoring , 2013 .

[17]  Manoj Kumar Tiwari,et al.  Kernel distance-based robust support vector methods and its application in developing a robust K-chart , 2006 .

[18]  Thomas E. Marlin,et al.  Multivariate statistical monitoring of process operating performance , 1991 .

[19]  Fugee Tsung,et al.  A kernel-distance-based multivariate control chart using support vector methods , 2003 .

[20]  Xu Wenli,et al.  Batch process monitoring based on functional data analysis and support vector data description , 2014 .

[21]  Robert P. W. Duin,et al.  Support vector domain description , 1999, Pattern Recognit. Lett..

[22]  Zhiqiang Ge,et al.  Improved two-level monitoring system for plant-wide processes , 2014 .

[23]  E. Parzen.  On Estimation of a Probability Density Function and Mode , 1962 .

[24]  Axel Coussement,et al.  Kernel density weighted principal component analysis of combustion processes , 2012 .

[25]  Chun-Chin Hsu,et al.  Intelligent ICA-SVM fault detector for non-Gaussian multivariate process monitoring , 2010, Expert Syst. Appl..

[26]  Fugee Tsung,et al.  Improved design of kernel distance–based charts using support vector methods , 2013 .

[27]  Claus Weihs,et al.  Kernel k-means clustering based local support vector domain description fault detection of multimodal processes , 2012, Expert Syst. Appl..

[28]  Xuefeng Yan,et al.  Probabilistic Weighted NPE-SVDD for chemical process monitoring , 2014 .

[29]  Ping Zhang,et al.  A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process , 2012 .

[30]  Steven X. Ding,et al.  A Survey of Fault Diagnosis and Fault-Tolerant Techniques—Part II: Fault Diagnosis With Knowledge-Based and Hybrid/Active Approaches , 2015, IEEE Transactions on Industrial Electronics.

[31]  Jie Yu,et al.  Quality relevant nonlinear batch process performance monitoring using a kernel based multiway non-Gaussian latent subspace projection approach , 2014 .

[32]  John F. MacGregor,et al.  Multi-way partial least squares in monitoring batch processes , 1995 .

[33]  R. Brereton,et al.  One class classifiers for process monitoring illustrated by the application to online HPLC of a continuous process , 2010 .

[34]  Furong Gao,et al.  Batch process monitoring based on support vector data description method , 2011 .

[35]  Theodora Kourti,et al.  Process analysis, monitoring and diagnosis, using multivariate projection methods , 1995 .

[36]  Zhiqiang Ge,et al.  Bagging support vector data description model for batch process monitoring , 2013 .