Documents disponibles écrits par cet auteur

Global local structure analysis model and its application for fault detection and identification / Muguang Zhang in Industrial & engineering chemistry research, Vol. 50 N° 11 (Juin 2011) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 50 N° 11 (Juin 2011) . - pp. 6837-6848 Titre : Global local structure analysis model and its application for fault detection and identification Type de document : texte imprimé Auteurs : Muguang Zhang, Auteur ; Zhiqiang Ge, Auteur ; Zhihuan Song, Auteur Année de publication : 2011 Article en page(s) : pp. 6837-6848 Note générale : Chimie industrielle Langues : Anglais (eng) Mots-clés : Failure  detection  Modeling Résumé : In this paper, a new fault detection and identification scheme that is based on the global―Iocal structure analysis (GLSA) model is proposed. By exploiting the underlying geometrical manifold and simultaneously keeping the global data information, the GLSA model constructs a dual-objective optimization function for dimension reduction of the process dataset. It combines the advantages of both locality preserving projections (LPP) and principal component analysis (PCA), under a unified framework Meanwhile, GLSA can successfully avoid the singularity problem that may occur in LPP and shares the orthogonal property of the PCA method. In order to balance the two subobjectives (corresponding to global and local structure preservings), a tuning parameter is introduced, and an energy-function-based strategy is proposed to determine the value of the introduced tuning parameter. For the purpose of fault detection, two statistics are constructed, based on the GLSA model Furthermore, the Bayesian inference algorithm is introduced upon the two monitoring statistics for fault identification. Two case studies are provided to demonstrate the efficiencies of the GLSA model. DEWEY : 660 ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=24199901 [article] Global local structure analysis model and its application for fault detection and identification [texte imprimé] / Muguang Zhang, Auteur ; Zhiqiang Ge, Auteur ; Zhihuan Song, Auteur . - 2011 . - pp. 6837-6848. Chimie industrielle Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 50 N° 11 (Juin 2011) . - pp. 6837-6848 Mots-clés : Failure  detection  Modeling Résumé : In this paper, a new fault detection and identification scheme that is based on the global―Iocal structure analysis (GLSA) model is proposed. By exploiting the underlying geometrical manifold and simultaneously keeping the global data information, the GLSA model constructs a dual-objective optimization function for dimension reduction of the process dataset. It combines the advantages of both locality preserving projections (LPP) and principal component analysis (PCA), under a unified framework Meanwhile, GLSA can successfully avoid the singularity problem that may occur in LPP and shares the orthogonal property of the PCA method. In order to balance the two subobjectives (corresponding to global and local structure preservings), a tuning parameter is introduced, and an energy-function-based strategy is proposed to determine the value of the introduced tuning parameter. For the purpose of fault detection, two statistics are constructed, based on the GLSA model Furthermore, the Bayesian inference algorithm is introduced upon the two monitoring statistics for fault identification. Two case studies are provided to demonstrate the efficiencies of the GLSA model. DEWEY : 660 ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=24199901

Kernel generalization of PPCA for nonlinear probabilistic monitoring / Ge, Zhiqiang 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.11832–11836 Titre : Kernel generalization of PPCA for nonlinear probabilistic monitoring Type de document : texte imprimé Auteurs : Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur Année de publication : 2011 Article en page(s) : pp.11832–11836 Note générale : Chimie industrielle Langues : Anglais (eng) Mots-clés : Surveillance Résumé : For probabilistic monitoring of nonlinear processes, the traditional probabilistic principal component analysis (PPCA)-based monitoring method is generalized through the kernel method. Thus, a probabilistic kernel PCA method is proposed for process monitoring in the present paper. Different from the traditional PPCA method, the new approach can successfully extract the nonlinear relationship between process variables. On the basis of the proposed nonlinear probabilistic monitoring approach, the monitoring performance of nonlinear processes can be effectively improved. To demonstrate the feasibility and efficiency of the proposed method, a case study on the Tennessee Eastman (TE) benchmark process is provided. DEWEY : 660 ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=23437886 [article] Kernel generalization of PPCA for nonlinear probabilistic monitoring [texte imprimé] / Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur . - 2011 . - pp.11832–11836. Chimie industrielle Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 22 (Novembre 2010) . - pp.11832–11836 Mots-clés : Surveillance Résumé : For probabilistic monitoring of nonlinear processes, the traditional probabilistic principal component analysis (PPCA)-based monitoring method is generalized through the kernel method. Thus, a probabilistic kernel PCA method is proposed for process monitoring in the present paper. Different from the traditional PPCA method, the new approach can successfully extract the nonlinear relationship between process variables. On the basis of the proposed nonlinear probabilistic monitoring approach, the monitoring performance of nonlinear processes can be effectively improved. To demonstrate the feasibility and efficiency of the proposed method, a case study on the Tennessee Eastman (TE) benchmark process is provided. DEWEY : 660 ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=23437886

A nonlinear probabilistic method for process monitoring / Ge, Zhiqiang in Industrial & engineering chemistry research, Vol. 49 N° 4 (Fevrier 2010) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 49 N° 4 (Fevrier 2010) . - pp 1770–1778 Titre : A nonlinear probabilistic method for process monitoring Type de document : texte imprimé Auteurs : Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur Année de publication : 2010 Article en page(s) : pp 1770–1778 Note générale : Chimie industrielle Langues : Anglais (eng) Mots-clés : Nonlinear  probabilistic  Monitoring  nonlinear. Résumé : To improve monitoring performance, the traditional principal component analysis (PCA) based process monitoring approach has been extended to its probabilistic counterpart. However, its ability is limited in linear processes. This paper proposes a nonlinear probabilistic method for monitoring nonlinear processes, which is based on generative topographic mapping (GTM). Similar to traditional methods, the monitoring statistic and its corresponding fault diagnosis approach have both been developed. Two case studies are provided to evaluate the feasibility and efficiency of the proposed method. DEWEY : 660 ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie900858v [article] A nonlinear probabilistic method for process monitoring [texte imprimé] / Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur . - 2010 . - pp 1770–1778. Chimie industrielle Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 4 (Fevrier 2010) . - pp 1770–1778 Mots-clés : Nonlinear  probabilistic  Monitoring  nonlinear. Résumé : To improve monitoring performance, the traditional principal component analysis (PCA) based process monitoring approach has been extended to its probabilistic counterpart. However, its ability is limited in linear processes. This paper proposes a nonlinear probabilistic method for monitoring nonlinear processes, which is based on generative topographic mapping (GTM). Similar to traditional methods, the monitoring statistic and its corresponding fault diagnosis approach have both been developed. Two case studies are provided to evaluate the feasibility and efficiency of the proposed method. DEWEY : 660 ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie900858v

Nonlinear probabilistic monitoring based on the gaussian process latent variable model / Ge, Zhiqiang in Industrial & engineering chemistry research, Vol. 49 N° 10 (Mai 2010) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 49 N° 10 (Mai 2010) . - pp. 4792–4799 Titre : Nonlinear probabilistic monitoring based on the gaussian process latent variable model Type de document : texte imprimé Auteurs : Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur Année de publication : 2010 Article en page(s) : pp. 4792–4799 Note générale : Industrial chemistry Langues : Anglais (eng) Mots-clés : Nonlinear  probabilistic  Gaussian  process Résumé : For probabilistic interpretation and monitoring performance enhancement in noisy processes, the probabilistic principal component analysis (PPCA) method has recently been introduced into the monitoring area. However, PPCA is restricted in linear processes. This paper first gives a new interpretation of PPCA through the Gaussian process manner. Then a new nonlinear probabilistic monitoring method is proposed, which is developed upon the Gaussian process latent variable model. Different from the traditional PPCA method, the new approach can successfully extract the nonlinear relationship between process variables. Furthermore, it exhibits more detailed information of uncertainty for process data, through which the operation condition and the fault behavior can be interpreted more easily. Two case studies are provided to show the efficiency of the proposed method. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie9019402 [article] Nonlinear probabilistic monitoring based on the gaussian process latent variable model [texte imprimé] / Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur . - 2010 . - pp. 4792–4799. Industrial chemistry Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 10 (Mai 2010) . - pp. 4792–4799 Mots-clés : Nonlinear  probabilistic  Gaussian  process Résumé : For probabilistic interpretation and monitoring performance enhancement in noisy processes, the probabilistic principal component analysis (PPCA) method has recently been introduced into the monitoring area. However, PPCA is restricted in linear processes. This paper first gives a new interpretation of PPCA through the Gaussian process manner. Then a new nonlinear probabilistic monitoring method is proposed, which is developed upon the Gaussian process latent variable model. Different from the traditional PPCA method, the new approach can successfully extract the nonlinear relationship between process variables. Furthermore, it exhibits more detailed information of uncertainty for process data, through which the operation condition and the fault behavior can be interpreted more easily. Two case studies are provided to show the efficiency of the proposed method. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie9019402

Nonlinear soft sensor development based on relevance vector machine / Ge, Zhiqiang in Industrial & engineering chemistry research, Vol. 49 N° 18 (Septembre 2010) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 49 N° 18 (Septembre 2010) . - pp. 8685–8693 Titre : Nonlinear soft sensor development based on relevance vector machine Type de document : texte imprimé Auteurs : Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur Année de publication : 2010 Article en page(s) : pp. 8685–8693 Note générale : Industrial chemistry Langues : Anglais (eng) Mots-clés : Nonlinear  soft  sensor Résumé : This paper proposes an effective nonlinear soft sensor based on relevance vector machine (RVM), which was originally proposed in the machine learning area. Compared to the widely used support vector machine (SVM) and least-squares support vector machine (LSSVM) based soft sensors, RVM gives a more sparse model structure, which can greatly reduce computational complexity for online prediction. While SVM/LSSVM can only provide a point estimation of the prediction result, RVM gives a probabilistic prediction result, which is more sophisticated for the soft sensor application. Furthermore, RVM can successfully avoid several drawbacks of the traditional support vector machine type method, such as kernel function limitation, parameter tuning complexity, and etc. Due to the advantages of RVM, a practical application of this method is made for soft sensor modeling in this paper. To evaluate the performance of the developed soft sensor, two case studies are demonstrated, which both support that RVM performs much better than other methods for soft sensing. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie101146d [article] Nonlinear soft sensor development based on relevance vector machine [texte imprimé] / Ge, Zhiqiang, Auteur ; Zhihuan Song, Auteur . - 2010 . - pp. 8685–8693. Industrial chemistry Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 18 (Septembre 2010) . - pp. 8685–8693 Mots-clés : Nonlinear  soft  sensor Résumé : This paper proposes an effective nonlinear soft sensor based on relevance vector machine (RVM), which was originally proposed in the machine learning area. Compared to the widely used support vector machine (SVM) and least-squares support vector machine (LSSVM) based soft sensors, RVM gives a more sparse model structure, which can greatly reduce computational complexity for online prediction. While SVM/LSSVM can only provide a point estimation of the prediction result, RVM gives a probabilistic prediction result, which is more sophisticated for the soft sensor application. Furthermore, RVM can successfully avoid several drawbacks of the traditional support vector machine type method, such as kernel function limitation, parameter tuning complexity, and etc. Due to the advantages of RVM, a practical application of this method is made for soft sensor modeling in this paper. To evaluate the performance of the developed soft sensor, two case studies are demonstrated, which both support that RVM performs much better than other methods for soft sensing. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie101146d

Quantitative analysis of influences of model plant mismatch on control loop behavior / Hong Wang in Industrial & engineering chemistry research, Vol. 51 N° 49 (Décembre 2012) 

Permalink

Robust online monitoring for multimode processes based on nonlinear external analysis / Ge, Zhiqiang in Industrial & engineering chemistry research, Vol. 47 n°14 (Juillet 2008) 

Permalink

Statistical prediction of product quality in batch processes with limited batch - cycle data / Ge, Zhiqiang in Industrial & engineering chemistry research, Vol. 51 N° 35 (Septembre 2012) 

Permalink

Transition process modeling and monitoring based on dynamic ensemble clustering and multiclass support vector data description / Zhibo Zhu in Industrial & engineering chemistry research, Vol. 50 N° 24 (Décembre 2011) 

Permalink