Adaptive soft sensor for online prediction and process monitoring based on a mixture of Gaussian process models. Grbić, R., Slišković, D., & Kadlec, P. Computers & Chemical Engineering, 58:84–97, November, 2013. ![Adaptive soft sensor for online prediction and process monitoring based on a mixture of Gaussian process models [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Linear models can be inappropriate when dealing with nonlinear and multimode processes, leading to a soft sensor with poor performance. Due to time-varying process behaviour it is necessary to derive and implement some kind of adaptation mechanism in order to keep the soft sensor performance at a desired level. Therefore, an adaptation mechanism for a soft sensor based on a mixture of Gaussian process regression models is proposed in this paper. A procedure for input variable selection based on mutual information is also presented. This procedure selects the most important input variables for output variable prediction, thus simplifying model development and adaptation. Apart from online prediction of the difficult-to-measure variable, this soft sensor can be used for adaptive process monitoring. The efficiency of the proposed method is benchmarked with the commonly applied recursive PLS and recursive PCA method on the Tennessee Eastman process and two real industrial examples.
@article{grbic_adaptive_2013,
title = {Adaptive soft sensor for online prediction and process monitoring based on a mixture of {Gaussian} process models},
volume = {58},
issn = {0098-1354},
url = {https://www.sciencedirect.com/science/article/pii/S0098135413002081},
doi = {10.1016/j.compchemeng.2013.06.014},
abstract = {Linear models can be inappropriate when dealing with nonlinear and multimode processes, leading to a soft sensor with poor performance. Due to time-varying process behaviour it is necessary to derive and implement some kind of adaptation mechanism in order to keep the soft sensor performance at a desired level. Therefore, an adaptation mechanism for a soft sensor based on a mixture of Gaussian process regression models is proposed in this paper. A procedure for input variable selection based on mutual information is also presented. This procedure selects the most important input variables for output variable prediction, thus simplifying model development and adaptation. Apart from online prediction of the difficult-to-measure variable, this soft sensor can be used for adaptive process monitoring. The efficiency of the proposed method is benchmarked with the commonly applied recursive PLS and recursive PCA method on the Tennessee Eastman process and two real industrial examples.},
language = {en},
urldate = {2022-05-02},
journal = {Computers \& Chemical Engineering},
author = {Grbić, Ratko and Slišković, Dražen and Kadlec, Petr},
month = nov,
year = {2013},
keywords = {Adaptive soft sensor, Gaussian process regression, Mutual information, Online prediction, Process modelling, Process monitoring},
pages = {84--97},
}
Downloads: 0
{"_id":"TcKGkxGCre7nkwCEd","bibbaseid":"grbi-slikovi-kadlec-adaptivesoftsensorforonlinepredictionandprocessmonitoringbasedonamixtureofgaussianprocessmodels-2013","author_short":["Grbić, R.","Slišković, D.","Kadlec, P."],"bibdata":{"bibtype":"article","type":"article","title":"Adaptive soft sensor for online prediction and process monitoring based on a mixture of Gaussian process models","volume":"58","issn":"0098-1354","url":"https://www.sciencedirect.com/science/article/pii/S0098135413002081","doi":"10.1016/j.compchemeng.2013.06.014","abstract":"Linear models can be inappropriate when dealing with nonlinear and multimode processes, leading to a soft sensor with poor performance. Due to time-varying process behaviour it is necessary to derive and implement some kind of adaptation mechanism in order to keep the soft sensor performance at a desired level. Therefore, an adaptation mechanism for a soft sensor based on a mixture of Gaussian process regression models is proposed in this paper. A procedure for input variable selection based on mutual information is also presented. This procedure selects the most important input variables for output variable prediction, thus simplifying model development and adaptation. Apart from online prediction of the difficult-to-measure variable, this soft sensor can be used for adaptive process monitoring. The efficiency of the proposed method is benchmarked with the commonly applied recursive PLS and recursive PCA method on the Tennessee Eastman process and two real industrial examples.","language":"en","urldate":"2022-05-02","journal":"Computers & Chemical Engineering","author":[{"propositions":[],"lastnames":["Grbić"],"firstnames":["Ratko"],"suffixes":[]},{"propositions":[],"lastnames":["Slišković"],"firstnames":["Dražen"],"suffixes":[]},{"propositions":[],"lastnames":["Kadlec"],"firstnames":["Petr"],"suffixes":[]}],"month":"November","year":"2013","keywords":"Adaptive soft sensor, Gaussian process regression, Mutual information, Online prediction, Process modelling, Process monitoring","pages":"84–97","bibtex":"@article{grbic_adaptive_2013,\n\ttitle = {Adaptive soft sensor for online prediction and process monitoring based on a mixture of {Gaussian} process models},\n\tvolume = {58},\n\tissn = {0098-1354},\n\turl = {https://www.sciencedirect.com/science/article/pii/S0098135413002081},\n\tdoi = {10.1016/j.compchemeng.2013.06.014},\n\tabstract = {Linear models can be inappropriate when dealing with nonlinear and multimode processes, leading to a soft sensor with poor performance. Due to time-varying process behaviour it is necessary to derive and implement some kind of adaptation mechanism in order to keep the soft sensor performance at a desired level. Therefore, an adaptation mechanism for a soft sensor based on a mixture of Gaussian process regression models is proposed in this paper. A procedure for input variable selection based on mutual information is also presented. This procedure selects the most important input variables for output variable prediction, thus simplifying model development and adaptation. Apart from online prediction of the difficult-to-measure variable, this soft sensor can be used for adaptive process monitoring. The efficiency of the proposed method is benchmarked with the commonly applied recursive PLS and recursive PCA method on the Tennessee Eastman process and two real industrial examples.},\n\tlanguage = {en},\n\turldate = {2022-05-02},\n\tjournal = {Computers \\& Chemical Engineering},\n\tauthor = {Grbić, Ratko and Slišković, Dražen and Kadlec, Petr},\n\tmonth = nov,\n\tyear = {2013},\n\tkeywords = {Adaptive soft sensor, Gaussian process regression, Mutual information, Online prediction, Process modelling, Process monitoring},\n\tpages = {84--97},\n}\n\n\n\n","author_short":["Grbić, R.","Slišković, D.","Kadlec, P."],"key":"grbic_adaptive_2013","id":"grbic_adaptive_2013","bibbaseid":"grbi-slikovi-kadlec-adaptivesoftsensorforonlinepredictionandprocessmonitoringbasedonamixtureofgaussianprocessmodels-2013","role":"author","urls":{"Paper":"https://www.sciencedirect.com/science/article/pii/S0098135413002081"},"keyword":["Adaptive soft sensor","Gaussian process regression","Mutual information","Online prediction","Process modelling","Process monitoring"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/mh_lenguyen","dataSources":["iwKepCrWBps7ojhDx"],"keywords":["adaptive soft sensor","gaussian process regression","mutual information","online prediction","process modelling","process monitoring"],"search_terms":["adaptive","soft","sensor","online","prediction","process","monitoring","based","mixture","gaussian","process","models","grbić","slišković","kadlec"],"title":"Adaptive soft sensor for online prediction and process monitoring based on a mixture of Gaussian process models","year":2013}