Documents disponibles écrits par cet auteur

Development of self - validating soft sensors using fast moving window partial least squares / Jialin Liu in Industrial & engineering chemistry research, Vol. 49 N° 22 (Novembre 2010) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 49 N° 22 (Novembre 2010) . - pp. 11530-11546 Titre : Development of self - validating soft sensors using fast moving window partial least squares Type de document : texte imprimé Auteurs : Jialin Liu, Auteur ; Ding-Sou Chen, Auteur ; Jui-Fu Shen, Auteur Année de publication : 2011 Article en page(s) : pp. 11530-11546 Note générale : Chimie industrielle Langues : Anglais (eng) Mots-clés : Software  sensor  Partial  least  squares Résumé : In the development of soft sensors for an industrial process, the colinearity of the predictor variables and the time-varying nature of the process need to be addressed. In many industrial applications, the partial least-squares (PLS) has been proven to capture the linear relationship between input and output variables for a local operating region; therefore, the PLS model needs to be adapted to accommodate the time-varying nature of the process. In this paper, a fast moving window algorithm is derived to update the PLS model. The proposed approach adapted the parameters of the inferential model with the dissimilarities between the new and oldest data and incorporated them into the kernel algorithm for the PLS. The computational loading of the model adaptation was therefore independent of the window size. In addition, the prediction performance of the model is only dependent on the retained latent variables (LVs) and the window size that can be predetermined from the historical data. Since a moving window approach is sensitive to outliers, the confidence intervals for the primary variables were created based on the prediction uncertainty. The inferential model would not be misled by the outliers from the online analyzers, whereas the model could be updated during the transition stage. The prediction performance of a soft sensor is not only dependent on the capability of the inferential model, but also relies on the data quality of the input measurements. In this paper, the input sensors were validated before performing a prediction. The deterioration of the prediction performance due to the failed sensors was removed by the reconstruction approach. A simulated example of a continuous stirred tank reactor (CSTR) with feedback control systems illustrated that the process characteristics captured by the PLS could be adapted to accommodate a nonlinear process. An industrial example, predicting oxygen concentrations in the air separation process, demonstrated the effectiveness of the proposed approach for the process industry. DEWEY : 660 ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=23437851 [article] Development of self - validating soft sensors using fast moving window partial least squares [texte imprimé] / Jialin Liu, Auteur ; Ding-Sou Chen, Auteur ; Jui-Fu Shen, Auteur . - 2011 . - pp. 11530-11546. Chimie industrielle Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 22 (Novembre 2010) . - pp. 11530-11546 Mots-clés : Software  sensor  Partial  least  squares Résumé : In the development of soft sensors for an industrial process, the colinearity of the predictor variables and the time-varying nature of the process need to be addressed. In many industrial applications, the partial least-squares (PLS) has been proven to capture the linear relationship between input and output variables for a local operating region; therefore, the PLS model needs to be adapted to accommodate the time-varying nature of the process. In this paper, a fast moving window algorithm is derived to update the PLS model. The proposed approach adapted the parameters of the inferential model with the dissimilarities between the new and oldest data and incorporated them into the kernel algorithm for the PLS. The computational loading of the model adaptation was therefore independent of the window size. In addition, the prediction performance of the model is only dependent on the retained latent variables (LVs) and the window size that can be predetermined from the historical data. Since a moving window approach is sensitive to outliers, the confidence intervals for the primary variables were created based on the prediction uncertainty. The inferential model would not be misled by the outliers from the online analyzers, whereas the model could be updated during the transition stage. The prediction performance of a soft sensor is not only dependent on the capability of the inferential model, but also relies on the data quality of the input measurements. In this paper, the input sensors were validated before performing a prediction. The deterioration of the prediction performance due to the failed sensors was removed by the reconstruction approach. A simulated example of a continuous stirred tank reactor (CSTR) with feedback control systems illustrated that the process characteristics captured by the PLS could be adapted to accommodate a nonlinear process. An industrial example, predicting oxygen concentrations in the air separation process, demonstrated the effectiveness of the proposed approach for the process industry. DEWEY : 660 ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=23437851

Fault detection and classification for a process with multiple production grades / Jialin Liu in Industrial & engineering chemistry research, Vol. 47 n°21 (Novembre 2008) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 47 n°21 (Novembre 2008) . - p. 8250–8262 Titre : Fault detection and classification for a process with multiple production grades Type de document : texte imprimé Auteurs : Jialin Liu, Auteur Année de publication : 2008 Article en page(s) : p. 8250–8262 Note générale : Chemical engineering Langues : Anglais (eng) Mots-clés : Industrial  polyethylene Résumé : In practice, an industrial polyethylene process produces various products, even often developing new production grades for market demand. Therefore, the process has not only multiple operating conditions, but also time-varying characteristics. In addition, the process measurements inevitably are redundant and noisy. It is a challenging problem for on-line classifying the operating conditions in the industrial process. In this paper, principal component analysis (PCA) is applied to the reference data set to reduce the dimensions of variables and eliminate the collinearities among process measurements. Since outliers are inevitable in a real plant data set, they significantly stretch the cluster centers and covariances and reach an unreliable solution. In this paper, the distance-based fuzzy c-means (DFCM) algorithm is proposed. A boundary distance for each cluster is derived for identifying outliers, which should be discarded from the reference data set. Before the on-line classification, the statistic Q and T2 of new data have to be evaluated. If any one of the statistics is out of its control limits, it indicates the new data do not belong to the PCA subspace and they should be collected for the next model update. In this paper, the blockwise recursive formulas for updating the means and covariance matrix are derived. By utilizing the updated means and covariance, a new PCA subspace that accounts for all events is derived recursively. In addition, through rotating and shifting the coordinates of the PCA subspace, the cluster parameters on the new subspace can be directly transferred from the previous one. The proposed method was successfully applied to monitor a polyethylene process with multiple production grades and time-varying characteristics. En ligne : http://pubs.acs.org/doi/abs/10.1021/ie0710014 [article] Fault detection and classification for a process with multiple production grades [texte imprimé] / Jialin Liu, Auteur . - 2008 . - p. 8250–8262. Chemical engineering Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 47 n°21 (Novembre 2008) . - p. 8250–8262 Mots-clés : Industrial  polyethylene Résumé : In practice, an industrial polyethylene process produces various products, even often developing new production grades for market demand. Therefore, the process has not only multiple operating conditions, but also time-varying characteristics. In addition, the process measurements inevitably are redundant and noisy. It is a challenging problem for on-line classifying the operating conditions in the industrial process. In this paper, principal component analysis (PCA) is applied to the reference data set to reduce the dimensions of variables and eliminate the collinearities among process measurements. Since outliers are inevitable in a real plant data set, they significantly stretch the cluster centers and covariances and reach an unreliable solution. In this paper, the distance-based fuzzy c-means (DFCM) algorithm is proposed. A boundary distance for each cluster is derived for identifying outliers, which should be discarded from the reference data set. Before the on-line classification, the statistic Q and T2 of new data have to be evaluated. If any one of the statistics is out of its control limits, it indicates the new data do not belong to the PCA subspace and they should be collected for the next model update. In this paper, the blockwise recursive formulas for updating the means and covariance matrix are derived. By utilizing the updated means and covariance, a new PCA subspace that accounts for all events is derived recursively. In addition, through rotating and shifting the coordinates of the PCA subspace, the cluster parameters on the new subspace can be directly transferred from the previous one. The proposed method was successfully applied to monitor a polyethylene process with multiple production grades and time-varying characteristics. En ligne : http://pubs.acs.org/doi/abs/10.1021/ie0710014

Fault detection and identification using modified bayesian classification on PCA subspace / Jialin Liu in Industrial & engineering chemistry research, Vol. 48 N° 6 (Mars 2009) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 48 N° 6 (Mars 2009) . - pp. 3059–3077 Titre : Fault detection and identification using modified bayesian classification on PCA subspace Type de document : texte imprimé Auteurs : Jialin Liu, Auteur ; Ding-Sou Chen, Auteur Année de publication : 2009 Article en page(s) : pp. 3059–3077 Note générale : Chemical engineering Langues : Anglais (eng) Mots-clés : Bayesian  classification  Monitoring  method  Principal  component  analysis  subspace Résumé : A novel process monitoring method based on modified Bayesian classification on PCA subspace is proposed. Fault detection and identification are the major steps to diagnose root causes of a process fault. However, before the faulty variables from the abnormal operations are identified, the different operating states need to be clustered from the historical data. The proposed approach modifies the Bayesian classification method to cluster data into groups. Therefore, a new fault identification index is derived based on cluster center and covariance. An industrial compressor process is used to demonstrate the effectiveness of the proposed approach. In the example, process-insight-based variables were monitored along with the measured variables. The capability of fault diagnosis has been improved, since the fault identification indices are directly related to the variables with process characteristics. En ligne : http://pubs.acs.org/doi/abs/10.1021/ie801243z [article] Fault detection and identification using modified bayesian classification on PCA subspace [texte imprimé] / Jialin Liu, Auteur ; Ding-Sou Chen, Auteur . - 2009 . - pp. 3059–3077. Chemical engineering Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 48 N° 6 (Mars 2009) . - pp. 3059–3077 Mots-clés : Bayesian  classification  Monitoring  method  Principal  component  analysis  subspace Résumé : A novel process monitoring method based on modified Bayesian classification on PCA subspace is proposed. Fault detection and identification are the major steps to diagnose root causes of a process fault. However, before the faulty variables from the abnormal operations are identified, the different operating states need to be clustered from the historical data. The proposed approach modifies the Bayesian classification method to cluster data into groups. Therefore, a new fault identification index is derived based on cluster center and covariance. An industrial compressor process is used to demonstrate the effectiveness of the proposed approach. In the example, process-insight-based variables were monitored along with the measured variables. The capability of fault diagnosis has been improved, since the fault identification indices are directly related to the variables with process characteristics. En ligne : http://pubs.acs.org/doi/abs/10.1021/ie801243z

Operational performance assessment and fault isolation for multimode processes / Jialin Liu in Industrial & engineering chemistry research, Vol. 49 N° 8 (Avril 2010) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 49 N° 8 (Avril 2010) . - pp. 3700–3714 Titre : Operational performance assessment and fault isolation for multimode processes Type de document : texte imprimé Auteurs : Jialin Liu, Auteur ; Ding-Sou Chen, Auteur Année de publication : 2010 Article en page(s) : pp. 3700–3714 Note générale : Industrial Chemistry Langues : Anglais (eng) Mots-clés : Operational  Performance  Assessment  Fault  Isolation  Multimode  Processes Résumé : Industrial processes are usually operated under normal conditions during daily production. For the purpose of improving product quality and productivity, the process engineers need to realize where the process variations come from. In this paper, according to the geometric shapes of clusters on the principal component (PC) subspace, the major variations along variable directions can be identified using a Pareto diagram and the 80/20 rule. In addition, fault isolation charts on the PC subspace are proposed to locate faulty variables when detecting abnormal events. A simulation example is demonstrated where the proposed approach is capable of locating the faulty variables without a smearing effect even in the case of multiple sensor faults. An industrial application of assessing operational performance and isolating faulty variables is presented. The results show that the faulty variables are highly correlated to the variables dominating the data variations in the normal operations, i.e., process-related abnormalities can be prevented by reducing the process variations during normal operations. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie901634r [article] Operational performance assessment and fault isolation for multimode processes [texte imprimé] / Jialin Liu, Auteur ; Ding-Sou Chen, Auteur . - 2010 . - pp. 3700–3714. Industrial Chemistry Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 8 (Avril 2010) . - pp. 3700–3714 Mots-clés : Operational  Performance  Assessment  Fault  Isolation  Multimode  Processes Résumé : Industrial processes are usually operated under normal conditions during daily production. For the purpose of improving product quality and productivity, the process engineers need to realize where the process variations come from. In this paper, according to the geometric shapes of clusters on the principal component (PC) subspace, the major variations along variable directions can be identified using a Pareto diagram and the 80/20 rule. In addition, fault isolation charts on the PC subspace are proposed to locate faulty variables when detecting abnormal events. A simulation example is demonstrated where the proposed approach is capable of locating the faulty variables without a smearing effect even in the case of multiple sensor faults. An industrial application of assessing operational performance and isolating faulty variables is presented. The results show that the faulty variables are highly correlated to the variables dominating the data variations in the normal operations, i.e., process-related abnormalities can be prevented by reducing the process variations during normal operations. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie901634r