Campaign-based modeling for degradation evolution in batch processes using a multiway partial least squares approach

Abstract In the process industry, various types of degradation occur in processing plants, resulting in significant economic losses. Modeling of degradation is important because it provides quantitative insights for consideration of degradation impacts in the operations of process manufacturing. This paper studies batch processes that show a periodic pattern for the evolution of degradation. A new data structure, the campaign, is applied for data-driven modeling of the periodic batch-to-batch evolution of degradation using a new multiway partial least squares approach, and it is further employed to predict the evolution of degradation in a series of batch runs. The proposed approach is illustrated and applied in a comprehensive industrial case study. The example illustrates the efficacy of the proposed model and presents a fair potential for applications of degradation prediction.

[1]  Huijun Gao,et al.  Data-Based Techniques Focused on Modern Industry: An Overview , 2015, IEEE Transactions on Industrial Electronics.

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

[3]  Lars Imsland,et al.  Robust state estimation for fouling evolution in batch processes using the EM algorithm , 2019 .

[4]  John F. MacGregor,et al.  Multivariate monitoring of batch processes using batch‐to‐batch information , 2004 .

[5]  A. J. Morris,et al.  Performance monitoring of a multi-product semi-batch process , 2001 .

[6]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[7]  Lars Imsland,et al.  Short-term Scheduling of a Multipurpose Batch Plant Considering Degradation Effects , 2019 .

[8]  Marappagounder Ramasamy,et al.  Heat exchanger fouling model and preventive maintenance scheduling tool , 2007 .

[9]  J. E. Strutt,et al.  A review of mechanical reliability modelling in relation to failure mechanisms , 1983 .

[10]  A. Ferrer,et al.  Dealing with missing data in MSPC: several methods, different interpretations, some examples , 2002 .

[11]  P. A. Taylor,et al.  The impact of missing measurements on PCA and PLS prediction and monitoring applications , 2006 .

[12]  Rajagopalan Srinivasan,et al.  Multi-model based real-time final product quality control strategy for batch processes , 2009, Comput. Chem. Eng..

[13]  Cenk Undey,et al.  Machine-learning for biopharmaceutical batch process monitoring with limited data , 2018 .

[14]  J. Macgregor,et al.  Analysis of multiblock and hierarchical PCA and PLS models , 1998 .

[15]  Matteo Cicciotti,et al.  Operational optimization of networks of compressors considering condition-based maintenance , 2016, Comput. Chem. Eng..

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

[17]  Zheng Liu,et al.  Latent Variable Models and Big Data in the Process Industries , 2015 .

[18]  Iiro Harjunkoski,et al.  Optimization of multipurpose process plant operations: A multi-time-scale maintenance and production scheduling approach , 2017, Comput. Chem. Eng..

[19]  Lars Imsland,et al.  Data-driven degradation model for batch processes: a case study on heat exchanger fouling , 2018 .

[20]  Jung-Ryul Lee,et al.  Structural health monitoring for a wind turbine system: a review of damage detection methods , 2008 .

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

[22]  Inmaculada Arauzo,et al.  Monitoring and prediction of fouling in coal-fired utility boilers using neural networks , 2005 .

[23]  Svante Wold,et al.  Batch Process Modeling and MSPC , 2009 .

[24]  John F. MacGregor,et al.  Product Quality Control in Semibatch Reactors Using Midcourse Correction Policies , 1997 .

[25]  Nina F. Thornhill,et al.  Online Performance Monitoring of Industrial Compressors Using Meanline Modelling , 2014 .

[26]  Ying Liang,et al.  A review of experimental measurement and prediction models of crude oil fouling rate in crude refinery preheat trains , 2015 .

[27]  A. Zmitrowicz.  Wear patterns and laws of wear – a review , 2006 .

[28]  A. J. Morris,et al.  Estimation of impurity and fouling in batch polymerisation reactors through the application of neural networks , 1999 .

[29]  Yuan Yao,et al.  Multivariate statistical process monitoring of batch‐to‐batch startups , 2015 .

[30]  Sandro Macchietto,et al.  Integration of Optimal Cleaning Scheduling and Control of Heat Exchanger Networks Undergoing Fouling: Model and Formulation , 2018, Industrial & Engineering Chemistry Research.

[31]  P. A. Taylor,et al.  Missing data methods in PCA and PLS: Score calculations with incomplete observations , 1996 .

[32]  J. Macgregor,et al.  Control of Particle Size Distributions in Emulsion Semibatch Polymerization Using Mid-Course Correction Policies , 2002 .

[33]  B. L. Yeap,et al.  Mitigation of Crude Oil Refinery Heat Exchanger Fouling Through Retrofits Based on Thermo-Hydraulic Fouling Models , 2004 .

[34]  H. Abdi.  Partial least squares regression and projection on latent structure regression (PLS Regression) , 2010 .

[35]  Sebastian Engell,et al.  Demand Side Management Scheduling Formulation for a Steel Plant Considering Electrode Degradation , 2019, IFAC-PapersOnLine.

[36]  S. Wold.  Cross-Validatory Estimation of the Number of Components in Factor and Principal Components Models , 1978 .

[37]  Yong Sun,et al.  A review on degradation models in reliability analysis , 2010, WCE 2010.

[38]  David Wang,et al.  Robust Data-Driven Modeling Approach for Real-Time Final Product Quality Prediction in Batch Process Operation , 2011, IEEE Transactions on Industrial Informatics.