Specific index-related process monitoring using a two-step information extraction method

Specific index-related process monitoring covers a wide range of requirements from industrial production. At present, it is still a challenge to divide into the specific index-related information and the specific index-unrelated information more accurately. In this paper, a two-step information extraction method is presented to solve this problem. In the first step, the overall process variables are separated into a strongly related variable set and a weakly related variable set. In the second step, the two sets are further divided into a specific index-related subset and a specific index-unrelated subset, respectively. Two specific index-related subsets form a new subspace related to the specific index. Two specific index-unrelated subsets are combined into a specific index-unrelated subspace. Two T2 statistics are built in the specific index-related subspace and the specific index-unrelated subspace, respectively. Finally, the proposed method is applied for Tennessee Eastman (TE). The results indicate the effectiveness of the proposed method.

[1]  Rongrong Sun,et al.  Fault diagnosis of nonlinear process based on KCPLS reconstruction , 2015 .

[2]  Kaixiang Peng,et al.  Quality-related process monitoring for dynamic non-Gaussian batch process with multi-phase using a new data-driven method , 2016, Neurocomputing.

[3]  S. Joe Qin,et al.  Quality‐relevant and process‐relevant fault monitoring with concurrent projection to latent structures , 2013 .

[4]  Kaixiang Peng,et al.  A comparison and evaluation of key performance indicator-based multivariate statistics process monitoring approaches ☆ , 2015 .

[5]  Yingwei Zhang,et al.  Quality-related fault detection approach based on dynamic kernel partial least squares , 2016 .

[6]  Hongbo Shi,et al.  Key principal components with recursive local outlier factor for multimode chemical process monitoring , 2016 .

[7]  E. F. Vogel,et al.  A plant-wide industrial process control problem , 1993 .

[8]  Kaixiang Peng,et al.  A Quality-Based Nonlinear Fault Diagnosis Framework Focusing on Industrial Multimode Batch Processes , 2016, IEEE Transactions on Industrial Electronics.

[9]  Hongbo Shi,et al.  Multimode process monitoring using improved dynamic neighborhood preserving embedding , 2014 .

[10]  Torsten Jeinsch,et al.  Quality-Related Fault Detection in Industrial Multimode Dynamic Processes , 2014, IEEE Transactions on Industrial Electronics.

[11]  Donghua Zhou,et al.  Quality Relevant Data-Driven Modeling and Monitoring of Multivariate Dynamic Processes: The Dynamic T-PLS Approach , 2011, IEEE Transactions on Neural Networks.

[12]  Guang Wang,et al.  A Kernel Least Squares Based Approach for Nonlinear Quality-Related Fault Detection , 2017, IEEE Transactions on Industrial Electronics.

[13]  Qiang Liu,et al.  Quality-Relevant Monitoring and Diagnosis with Dynamic Concurrent Projection to Latent Structures , 2014 .

[14]  Xu Yang,et al.  Robust PLS approach for KPI-related prediction and diagnosis against outliers and missing data , 2014, Int. J. Syst. Sci..

[15]  Xuefeng Yan,et al.  Batch process monitoring based on multiple-phase online sorting principal component analysis. , 2016, ISA transactions.

[16]  Zhiqiang Ge,et al.  Supervised linear dynamic system model for quality related fault detection in dynamic processes , 2016 .

[17]  Huijun Gao,et al.  Data-Driven Process Monitoring Based on Modified Orthogonal Projections to Latent Structures , 2016, IEEE Transactions on Control Systems Technology.

[18]  Chunhui Zhao,et al.  Comprehensive Subspace Decomposition with Analysis of Between-Mode Relative Changes for Multimode Process Monitoring , 2015 .

[19]  Kaixiang Peng,et al.  Adaptive total PLS based quality-relevant process monitoring with application to the Tennessee Eastman process , 2015, Neurocomputing.

[20]  Donghua Zhou,et al.  Total projection to latent structures for process monitoring , 2009 .