An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning
Due to significant differences in data distribution under different working conditions during Pichia pastoris biochemical reaction process, traditional soft-sensor model suffer from the model failure and deterioration, this paper propose a soft-sensor modeling method combing long short-term memory n...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-10-01
|
Series: | Alexandria Engineering Journal |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016823007883 |
_version_ | 1797659912469217280 |
---|---|
author | Bo Wang Yongxin Nie Ligang Zhang Yongxian Song Qiwei Zhu |
author_facet | Bo Wang Yongxin Nie Ligang Zhang Yongxian Song Qiwei Zhu |
author_sort | Bo Wang |
collection | DOAJ |
description | Due to significant differences in data distribution under different working conditions during Pichia pastoris biochemical reaction process, traditional soft-sensor model suffer from the model failure and deterioration, this paper propose a soft-sensor modeling method combing long short-term memory network (LSTM) and balanced distribution adaptation method (BDA). Firstly, the source domain data is used to establish an accurate source domain LSTM prediction model, and the structure and parameters of the first layer of LSTM are fixed to migrate to the target domain prediction model. Then use the balanced distribution adaptation method to shrink the distribution differences between different domains of data. Finally, data that has been modeled with balanced and adaptive distribution assist the real-time data to train the remaining layer of the network, and the accurate target domain prediction model is finally obtained. The simulation results show that the mentioned method has the preponderance of timely prediction and high prediction accuracy, which validates the effectiveness and practicality of the method. This method solves the problem of soft-sensor modeling under unknown modes of multiple operating conditions in Pichia pastoris biochemical reaction process, achieving the prediction of key parameters under different operating conditions, which can be widely applied, and also providing a new method for soft-sensor modeling of other non system systems. |
first_indexed | 2024-03-11T18:22:06Z |
format | Article |
id | doaj.art-654412beb83d48f3b59fcdecd43d433c |
institution | Directory Open Access Journal |
issn | 1110-0168 |
language | English |
last_indexed | 2024-03-11T18:22:06Z |
publishDate | 2023-10-01 |
publisher | Elsevier |
record_format | Article |
series | Alexandria Engineering Journal |
spelling | doaj.art-654412beb83d48f3b59fcdecd43d433c2023-10-15T04:36:46ZengElsevierAlexandria Engineering Journal1110-01682023-10-0181170177An soft-sensor method for the biochemical reaction process based on LSTM and transfer learningBo Wang0Yongxin Nie1Ligang Zhang2Yongxian Song3Qiwei Zhu4School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, China; Corresponding author.School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, ChinaSchool of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, ChinaCollege of Electronic Engineering, Nanjing XiaoZhuang University, Nanjing 211171, ChinaSchool of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, ChinaDue to significant differences in data distribution under different working conditions during Pichia pastoris biochemical reaction process, traditional soft-sensor model suffer from the model failure and deterioration, this paper propose a soft-sensor modeling method combing long short-term memory network (LSTM) and balanced distribution adaptation method (BDA). Firstly, the source domain data is used to establish an accurate source domain LSTM prediction model, and the structure and parameters of the first layer of LSTM are fixed to migrate to the target domain prediction model. Then use the balanced distribution adaptation method to shrink the distribution differences between different domains of data. Finally, data that has been modeled with balanced and adaptive distribution assist the real-time data to train the remaining layer of the network, and the accurate target domain prediction model is finally obtained. The simulation results show that the mentioned method has the preponderance of timely prediction and high prediction accuracy, which validates the effectiveness and practicality of the method. This method solves the problem of soft-sensor modeling under unknown modes of multiple operating conditions in Pichia pastoris biochemical reaction process, achieving the prediction of key parameters under different operating conditions, which can be widely applied, and also providing a new method for soft-sensor modeling of other non system systems.http://www.sciencedirect.com/science/article/pii/S1110016823007883Long short-term memory networkTransfer learningPichia pastorisVariable working conditionsSoft-sensor |
spellingShingle | Bo Wang Yongxin Nie Ligang Zhang Yongxian Song Qiwei Zhu An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning Alexandria Engineering Journal Long short-term memory network Transfer learning Pichia pastoris Variable working conditions Soft-sensor |
title | An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning |
title_full | An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning |
title_fullStr | An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning |
title_full_unstemmed | An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning |
title_short | An soft-sensor method for the biochemical reaction process based on LSTM and transfer learning |
title_sort | soft sensor method for the biochemical reaction process based on lstm and transfer learning |
topic | Long short-term memory network Transfer learning Pichia pastoris Variable working conditions Soft-sensor |
url | http://www.sciencedirect.com/science/article/pii/S1110016823007883 |
work_keys_str_mv | AT bowang ansoftsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT yongxinnie ansoftsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT ligangzhang ansoftsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT yongxiansong ansoftsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT qiweizhu ansoftsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT bowang softsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT yongxinnie softsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT ligangzhang softsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT yongxiansong softsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning AT qiweizhu softsensormethodforthebiochemicalreactionprocessbasedonlstmandtransferlearning |