Data-Driven Process Monitoring Based on Modified Orthogonal Projections to Latent Structures

Quality- or output-related fault detection has attracted much attention in recent years. Several approaches have been developed to solve this issue based on postprocessing schemes. However, further studies find that these methods gradually lose their functions when amplitudes of quality-unrelated faults increase; in addition, they still consume a relatively large amount of calculation load in practice. In this brief, we propose a new structure of preprocessing-modeling-postprocessing, within which modified orthogonal projections to latent structures (MOPLS) method is developed. Compared with the previous approaches, the new method significantly improves the performance of quality-related fault detection. In addition, it reduces the number of required latent variables, thus it has a quite lower computational load than the previous ones. A numerical example and the Tennessee Eastman process are used to verify the effectiveness of the proposed approach.

[1]  Ali Cinar,et al.  Monitoring, fault diagnosis, fault-tolerant control and optimization: Data driven methods , 2012, Comput. Chem. Eng..

[2]  Donghua Zhou,et al.  Geometric properties of partial least squares for process monitoring , 2010, Autom..

[3]  Steven X. Ding,et al.  Study on modifications of PLS approach for process monitoring , 2011 .

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

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

[6]  S. Wold,et al.  Orthogonal projections to latent structures (O‐PLS) , 2002 .

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

[8]  Bhupinder S. Dayal,et al.  Improved PLS algorithms , 1997 .

[9]  Guang Wang,et al.  Quality-Related Fault Detection Approach Based on Orthogonal Signal Correction and Modified PLS , 2015, IEEE Transactions on Industrial Informatics.

[10]  Nola D. Tracy,et al.  Multivariate Control Charts for Individual Observations , 1992 .

[11]  S. Qin,et al.  Selection of the Number of Principal Components: The Variance of the Reconstruction Error Criterion with a Comparison to Other Methods† , 1999 .

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

[13]  Tiago J. Rato,et al.  Fault detection in the Tennessee Eastman benchmark process using dynamic principal components analysis based on decorrelated residuals (DPCA-DR) , 2013 .

[14]  Ronald R. Coifman,et al.  The prediction error in CLS and PLS: the importance of feature selection prior to multivariate calibration , 2005 .

[15]  N. Lawrence Ricker,et al.  Decentralized control of the Tennessee Eastman Challenge Process , 1996 .

[16]  Mark F. Davis,et al.  Orthogonal projection to latent structures solution properties for chemometrics and systems biology data , 2011 .