Documents disponibles écrits par cet auteur

Inferential model for industrial polypropylene melt index prediction with embedded priori knowledge and delay estimation / Haichuan Lou in Industrial & engineering chemistry research, Vol. 51 N° 25 (Juin 2012) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 51 N° 25 (Juin 2012) . - pp. 8510-8525 Titre : Inferential model for industrial polypropylene melt index prediction with embedded priori knowledge and delay estimation Type de document : texte imprimé Auteurs : Haichuan Lou, Auteur ; Hongye Su, Auteur ; Lei Xie, Auteur Année de publication : 2012 Article en page(s) : pp. 8510-8525 Note générale : Industrial chemistry Langues : Anglais (eng) Mots-clés : Prediction  Modeling Résumé : Melt index inferential model plays an important role in the control and optimization of polypropylene production. This study proposed a novel multiple-priori-knowledge based neural network (MPKNN) inferential model for melt index prediction. The prior knowledge from the industrial propylene polymerization process is fully exploited and embedded into the construction of multilayer perceptron neural network in the form of nonlinear constraints. Meanwhile, an adaptive PSO-SQP (particle swarm optimization-sequential quadratics programming) is proposed to optimize the network weights. The proposed MPKNN model has good fitting and prediction ability. Meanwhile, it can avoid unwanted zero value and wrong signal of the model gains. By embedding priori knowledge, the model ensures the safety in the quality control of melt index. In addition, a hybrid model combining the MPKNN model with a simplified mechanism model is proposed to enhance the extrapolation capability. A normalized mutual information method is employed to estimate the delay between independent variables and dependent variables. The proposed hybrid inferential model is validated on recorded data from an industrial double-loop propylene-polymerization reaction process. ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=26066778 [article] Inferential model for industrial polypropylene melt index prediction with embedded priori knowledge and delay estimation [texte imprimé] / Haichuan Lou, Auteur ; Hongye Su, Auteur ; Lei Xie, Auteur . - 2012 . - pp. 8510-8525. Industrial chemistry Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 51 N° 25 (Juin 2012) . - pp. 8510-8525 Mots-clés : Prediction  Modeling Résumé : Melt index inferential model plays an important role in the control and optimization of polypropylene production. This study proposed a novel multiple-priori-knowledge based neural network (MPKNN) inferential model for melt index prediction. The prior knowledge from the industrial propylene polymerization process is fully exploited and embedded into the construction of multilayer perceptron neural network in the form of nonlinear constraints. Meanwhile, an adaptive PSO-SQP (particle swarm optimization-sequential quadratics programming) is proposed to optimize the network weights. The proposed MPKNN model has good fitting and prediction ability. Meanwhile, it can avoid unwanted zero value and wrong signal of the model gains. By embedding priori knowledge, the model ensures the safety in the quality control of melt index. In addition, a hybrid model combining the MPKNN model with a simplified mechanism model is proposed to enhance the extrapolation capability. A normalized mutual information method is employed to estimate the delay between independent variables and dependent variables. The proposed hybrid inferential model is validated on recorded data from an industrial double-loop propylene-polymerization reaction process. ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=26066778

A Multiperiod optimization model for hydrogen system scheduling in refinery / Yunqiang Jiao in Industrial & engineering chemistry research, Vol. 51 N° 17 (Mai 2012) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 51 N° 17 (Mai 2012) . - pp. 6085–6098 Titre : A Multiperiod optimization model for hydrogen system scheduling in refinery Type de document : texte imprimé Auteurs : Yunqiang Jiao, Auteur ; Hongye Su, Auteur ; Weifeng Hou, Auteur Année de publication : 2012 Article en page(s) : pp. 6085–6098 Note générale : Industrial chemistry Langues : Anglais (eng) Mots-clés : Hydrogenation  process Résumé : In a refinery, hydrogen, as a valuable resource, is also a byproduct and a significant raw material source of the petroleum refining and petrochemical hydrogenation process. To reduce costs and save energy for the petrochemical industry, the hydrogen system in a refinery should be operated under the optimal scheme to meet the varying hydrogen demands of hydrogen consumers. Optimal scheduling of the hydrogen system can help a refinery to achieve cost reduction and cleaner production. In this paper, a discrete-time mixed-integer nonlinear programming (MINLP) model that considers the penalties for abnormal situations in the hydrogen pipe network (HPN), compressors start–stop, and changes in hydrogen sources for hydrogen consumers is proposed for the optimal scheduling of the hydrogen system under multiperiod operation. The solution of the scheduling problem is obtained based on an iterative method between that of a mixed-integer linear programming (MILP) problem and that of a nonlinear programming (NLP) problem, avoiding the solution of the MINLP problem directly and the occurrence of composition discrepancy. A case study based on the data from a real refinery is presented to illustrate the effectiveness and feasibility of the proposed methodology. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie2019239 [article] A Multiperiod optimization model for hydrogen system scheduling in refinery [texte imprimé] / Yunqiang Jiao, Auteur ; Hongye Su, Auteur ; Weifeng Hou, Auteur . - 2012 . - pp. 6085–6098. Industrial chemistry Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 51 N° 17 (Mai 2012) . - pp. 6085–6098 Mots-clés : Hydrogenation  process Résumé : In a refinery, hydrogen, as a valuable resource, is also a byproduct and a significant raw material source of the petroleum refining and petrochemical hydrogenation process. To reduce costs and save energy for the petrochemical industry, the hydrogen system in a refinery should be operated under the optimal scheme to meet the varying hydrogen demands of hydrogen consumers. Optimal scheduling of the hydrogen system can help a refinery to achieve cost reduction and cleaner production. In this paper, a discrete-time mixed-integer nonlinear programming (MINLP) model that considers the penalties for abnormal situations in the hydrogen pipe network (HPN), compressors start–stop, and changes in hydrogen sources for hydrogen consumers is proposed for the optimal scheduling of the hydrogen system under multiperiod operation. The solution of the scheduling problem is obtained based on an iterative method between that of a mixed-integer linear programming (MILP) problem and that of a nonlinear programming (NLP) problem, avoiding the solution of the MINLP problem directly and the occurrence of composition discrepancy. A case study based on the data from a real refinery is presented to illustrate the effectiveness and feasibility of the proposed methodology. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie2019239

Support vector regression approach for simultaneous data reconciliation and gross error or outlier detection / Yu Miao in Industrial & engineering chemistry research, Vol. 48 N° 24 (Décembre 2009) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 48 N° 24 (Décembre 2009) . - pp. 10903–10911 Titre : Support vector regression approach for simultaneous data reconciliation and gross error or outlier detection Type de document : texte imprimé Auteurs : Yu Miao, Auteur ; Hongye Su, Auteur ; Ouguan Xu, Auteur Année de publication : 2010 Article en page(s) : pp. 10903–10911 Note générale : Industrial chemistry Langues : Anglais (eng) Mots-clés : Support--Vector--Regression--Approach--Simultaneous--Data--Reconciliation--Gross  Error--Outlier  Detection Résumé : Process data measurements are important for model fitting, process monitoring, control, optimization, and management decision making, and they are usually applied with parameter estimation. This paper applies the support vector (SV) regression approach as a framework for simultaneous data reconciliation and gross error or outlier detection in processes. SV regression minimizes regularized risk instead of maximum likelihood, and it is a compromise between empirical risk and model complexity. For data reconciliation, it is robust to random and gross errors or outliers, because loss functions are addressed as objective functions instead of least-squares. Furthermore, because of the robust nature of SV regressions, the coefficients of the SV regression itself have less effect on the reconciled results. Finally, a number of process and literature system simulation results show the features of the SV regression approach for data reconciliation proposed in this paper. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie801629f [article] Support vector regression approach for simultaneous data reconciliation and gross error or outlier detection [texte imprimé] / Yu Miao, Auteur ; Hongye Su, Auteur ; Ouguan Xu, Auteur . - 2010 . - pp. 10903–10911. Industrial chemistry Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 48 N° 24 (Décembre 2009) . - pp. 10903–10911 Mots-clés : Support--Vector--Regression--Approach--Simultaneous--Data--Reconciliation--Gross  Error--Outlier  Detection Résumé : Process data measurements are important for model fitting, process monitoring, control, optimization, and management decision making, and they are usually applied with parameter estimation. This paper applies the support vector (SV) regression approach as a framework for simultaneous data reconciliation and gross error or outlier detection in processes. SV regression minimizes regularized risk instead of maximum likelihood, and it is a compromise between empirical risk and model complexity. For data reconciliation, it is robust to random and gross errors or outliers, because loss functions are addressed as objective functions instead of least-squares. Furthermore, because of the robust nature of SV regressions, the coefficients of the SV regression itself have less effect on the reconciled results. Finally, a number of process and literature system simulation results show the features of the SV regression approach for data reconciliation proposed in this paper. ISSN : 0888-5885 En ligne : http://pubs.acs.org/doi/abs/10.1021/ie801629f

Support vector regression approach for simultaneous data reconciliation and gross error or outlier detection / Yu Miao in Industrial & engineering chemistry research, Vol. 48 N° 24 (Décembre 2009) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 48 N° 24 (Décembre 2009) . - pp. 10903–10911 Titre : Support vector regression approach for simultaneous data reconciliation and gross error or outlier detection Type de document : texte imprimé Auteurs : Yu Miao, Auteur ; Hongye Su, Auteur ; Ouguan Xu, Auteur Année de publication : 2010 Article en page(s) : pp. 10903–10911 Note générale : Chemical engineering Langues : Anglais (eng) Mots-clés : Support  vector  regression  Outlier  detection Résumé : Process data measurements are important for model fitting, process monitoring, control, optimization, and management decision making, and they are usually applied with parameter estimation. This paper applies the support vector (SV) regression approach as a framework for simultaneous data reconciliation and gross error or outlier detection in processes. SV regression minimizes regularized risk instead of maximum likelihood, and it is a compromise between empirical risk and model complexity. For data reconciliation, it is robust to random and gross errors or outliers, because loss functions are addressed as objective functions instead of least-squares. Furthermore, because of the robust nature of SV regressions, the coefficients of the SV regression itself have less effect on the reconciled results. Finally, a number of process and literature system simulation results show the features of the SV regression approach for data reconciliation proposed in this paper. En ligne : http://pubs.acs.org/doi/abs/10.1021/ie801629f [article] Support vector regression approach for simultaneous data reconciliation and gross error or outlier detection [texte imprimé] / Yu Miao, Auteur ; Hongye Su, Auteur ; Ouguan Xu, Auteur . - 2010 . - pp. 10903–10911. Chemical engineering Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 48 N° 24 (Décembre 2009) . - pp. 10903–10911 Mots-clés : Support  vector  regression  Outlier  detection Résumé : Process data measurements are important for model fitting, process monitoring, control, optimization, and management decision making, and they are usually applied with parameter estimation. This paper applies the support vector (SV) regression approach as a framework for simultaneous data reconciliation and gross error or outlier detection in processes. SV regression minimizes regularized risk instead of maximum likelihood, and it is a compromise between empirical risk and model complexity. For data reconciliation, it is robust to random and gross errors or outliers, because loss functions are addressed as objective functions instead of least-squares. Furthermore, because of the robust nature of SV regressions, the coefficients of the SV regression itself have less effect on the reconciled results. Finally, a number of process and literature system simulation results show the features of the SV regression approach for data reconciliation proposed in this paper. En ligne : http://pubs.acs.org/doi/abs/10.1021/ie801629f