Supervised local and non-local structure preserving projections with application to just-in-time learning for adaptive soft sensor

Abstract In soft sensor field, just-in-time learning (JITL) is an effective approach to model nonlinear and time varying processes. However, most similarity criterions in JITL are computed in the input space only while ignoring important output information, which may lead to inaccurate construction of relevant sample set. To solve this problem, we propose a novel supervised feature extraction method suitable for the regression problem called supervised local and non-local structure preserving projections (SLNSPP), in which both input and output information can be easily and effectively incorporated through a newly defined similarity index. The SLNSPP can not only retain the virtue of locality preserving projections but also prevent faraway points from nearing after projection, which endues SLNSPP with powerful discriminating ability. Such two good properties of SLNSPP are desirable for JITL as they are expected to enhance the accuracy of similar sample selection. Consequently, we present a SLNSPP-JITL framework for developing adaptive soft sensor, including a sparse learning strategy to limit the scale and update the frequency of database. Finally, two case studies are conducted with benchmark datasets to evaluate the performance of the proposed schemes. The results demonstrate the effectiveness of LNSPP and SLNSPP.

[1]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machines , 2002 .

[2]  Zhi-huan Song,et al.  Adaptive local kernel-based learning for soft sensor modeling of nonlinear processes , 2011 .

[3]  Gavin C. Cawley,et al.  Preventing Over-Fitting during Model Selection via Bayesian Regularisation of the Hyper-Parameters , 2007, J. Mach. Learn. Res..

[4]  Honglong Chen,et al.  Adaptive Anti-Over-Fitting Soft Sensing Method Based on Local Learning , 2013 .

[5]  Hiromasa Kaneko,et al.  Database monitoring index for adaptive soft sensors and the application to industrial process , 2014 .

[6]  Xuemin Tian,et al.  Soft sensor for polypropylene melt index based on improved orthogonal least squares , 2010, 2010 8th World Congress on Intelligent Control and Automation.

[7]  Honggang Zhang,et al.  Comments on "Globally Maximizing, Locally Minimizing: Unsupervised Discriminant Projection with Application to Face and Palm Biometrics" , 2007, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Luigi Fortuna,et al.  Soft Sensors for Monitoring and Control of Industrial Processes (Advances in Industrial Control) , 2006 .

[9]  Johan A. K. Suykens,et al.  Weighted least squares support vector machines: robustness and sparse approximation , 2002, Neurocomputing.

[10]  Tianyou Chai,et al.  On-line KPLS Algorithm with Application to Ensemble Modeling Parameters of Mill Load: On-line KPLS Algorithm with Application to Ensemble Modeling Parameters of Mill Load , 2014 .

[11]  Min-Sen Chiu,et al.  Adaptive generalized predictive control based on JITL technique , 2009 .

[12]  Zengliang Gao,et al.  Just-in-time kernel learning with adaptive parameter selection for soft sensor modeling of batch processes , 2012 .

[13]  Yi Liu,et al.  Development of soft-sensors for online quality prediction of sequential-reactor-multi-grade industrial processes , 2013 .

[14]  Ping Li,et al.  Selective recursive kernel learning for online identification of nonlinear systems with NARX form , 2010 .

[15]  Hiromasa Kaneko,et al.  Application of Online Support Vector Regression for Soft Sensors , 2014 .

[16]  Manabu Kano,et al.  Soft‐sensor development using correlation‐based just‐in‐time modeling , 2009 .

[17]  Koichi Fujiwara,et al.  Development of correlation-based pattern recognition algorithm and adaptive soft-sensor design , 2012 .

[18]  Bogdan Gabrys,et al.  Data-driven Soft Sensors in the process industry , 2009, Comput. Chem. Eng..

[19]  Deli Zhao,et al.  Linear local tangent space alignment and application to face recognition , 2007, Neurocomputing.

[20]  Hare Krishna Mohanta,et al.  Development and comparison of neural network based soft sensors for online estimation of cement clinker quality. , 2013, ISA transactions.

[21]  Hidenori Kimura,et al.  Just‐in‐Time Modeling for Function Prediction and Its Applications , 2001 .

[22]  Zhi-huan Song,et al.  Global–Local Structure Analysis Model and Its Application for Fault Detection and Identification , 2011 .

[23]  Daoping Huang,et al.  Development of Interval Soft Sensors Using Enhanced Just-in-Time Learning and Inductive Confidence Predictor , 2012 .

[24]  Zhiqiang Ge,et al.  Semisupervised Bayesian method for soft sensor modeling with unlabeled data samples , 2011 .

[25]  Fuli Wang,et al.  Real-time product quality control for batch processes based on stacked least-squares support vector regression models , 2012, Comput. Chem. Eng..

[26]  Bogdan Gabrys,et al.  Review of adaptation mechanisms for data-driven soft sensors , 2011, Comput. Chem. Eng..

[27]  Maya R. Gupta,et al.  Adaptive Local Linear Regression With Application to Printer Color Management , 2008, IEEE Transactions on Image Processing.

[28]  Wang Ping.  Online learning soft sensor method based on recursive kernel algorithm for PLS , 2012 .

[29]  Jin Wang,et al.  Adaptive Outlier Detection and Classification for Online Soft Sensor Update , 2012 .

[30]  Xiaofei He,et al.  Locality Preserving Projections , 2003, NIPS.

[31]  Hiromasa Kaneko,et al.  Classification of the Degradation of Soft Sensor Models and Discussion on Adaptive Models , 2013 .

[32]  M. Chiu,et al.  A new data-based methodology for nonlinear process modeling , 2004 .

[33]  Tang Jian,et al.  On-line KPLS Algorithm with Application to Ensemble Modeling Parameters of Mill Load , 2013 .

[34]  Jianping Zhang,et al.  Intelligent Selection of Instances for Prediction Functions in Lazy Learning Algorithms , 1997, Artificial Intelligence Review.