Cargando…

Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation

[Image: see text] The microbial fermentation process often involves various biological metabolic reactions and chemical processes. The mixed bacterial culture process of 2-keto-l-gulonic acid has strong nonlinear and time-varying characteristics. In this study, a probabilistic Bayesian deep learning...

Descripción completa

Detalles Bibliográficos
Autores principales: Wang, Tao, You, Jiebing, Gong, Xiugang, Yang, Shanliang, Wang, Lei, Chang, Zheng
Formato: Online Artículo Texto
Lenguaje:English
Publicado: American Chemical Society 2023
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10357427/
https://www.ncbi.nlm.nih.gov/pubmed/37483241
http://dx.doi.org/10.1021/acsomega.3c02387
_version_ 1785075491376463872
author Wang, Tao
You, Jiebing
Gong, Xiugang
Yang, Shanliang
Wang, Lei
Chang, Zheng
author_facet Wang, Tao
You, Jiebing
Gong, Xiugang
Yang, Shanliang
Wang, Lei
Chang, Zheng
author_sort Wang, Tao
collection PubMed
description [Image: see text] The microbial fermentation process often involves various biological metabolic reactions and chemical processes. The mixed bacterial culture process of 2-keto-l-gulonic acid has strong nonlinear and time-varying characteristics. In this study, a probabilistic Bayesian deep learning approach is proposed to obtain a highly accurate and robust prediction of product formation. The Bayesian optimized deep neural network (BODNN) is utilized as basic model for prediction, the structural parameters of which are optimized. Then, the training datasets are classified into different categories according to the prior evaluation of prediction error. The final forecasting is a weighted combination of BODNN models based on the Bayesian hybrid method. The weights can be interpreted as Bayesian posterior probabilities and are computed recursively. The validation of 95 industrial batches is carried out, and the average root mean square errors are 1.51 and 2.01% for 4 and 8 h ahead prediction, respectively. The results illustrate that the proposed approach can capture the dynamics of fermentation batches and is suitable for online process monitoring.
format Online
Article
Text
id pubmed-10357427
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher American Chemical Society
record_format MEDLINE/PubMed
spelling pubmed-103574272023-07-21 Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation Wang, Tao You, Jiebing Gong, Xiugang Yang, Shanliang Wang, Lei Chang, Zheng ACS Omega [Image: see text] The microbial fermentation process often involves various biological metabolic reactions and chemical processes. The mixed bacterial culture process of 2-keto-l-gulonic acid has strong nonlinear and time-varying characteristics. In this study, a probabilistic Bayesian deep learning approach is proposed to obtain a highly accurate and robust prediction of product formation. The Bayesian optimized deep neural network (BODNN) is utilized as basic model for prediction, the structural parameters of which are optimized. Then, the training datasets are classified into different categories according to the prior evaluation of prediction error. The final forecasting is a weighted combination of BODNN models based on the Bayesian hybrid method. The weights can be interpreted as Bayesian posterior probabilities and are computed recursively. The validation of 95 industrial batches is carried out, and the average root mean square errors are 1.51 and 2.01% for 4 and 8 h ahead prediction, respectively. The results illustrate that the proposed approach can capture the dynamics of fermentation batches and is suitable for online process monitoring. American Chemical Society 2023-07-04 /pmc/articles/PMC10357427/ /pubmed/37483241 http://dx.doi.org/10.1021/acsomega.3c02387 Text en © 2023 The Authors. Published by American Chemical Society https://creativecommons.org/licenses/by-nc-nd/4.0/Permits non-commercial access and re-use, provided that author attribution and integrity are maintained; but does not permit creation of adaptations or other derivative works (https://creativecommons.org/licenses/by-nc-nd/4.0/).
spellingShingle Wang, Tao
You, Jiebing
Gong, Xiugang
Yang, Shanliang
Wang, Lei
Chang, Zheng
Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation
title Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation
title_full Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation
title_fullStr Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation
title_full_unstemmed Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation
title_short Probabilistic Bayesian Deep Learning Approach for Online Forecasting of Fed-Batch Fermentation
title_sort probabilistic bayesian deep learning approach for online forecasting of fed-batch fermentation
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10357427/
https://www.ncbi.nlm.nih.gov/pubmed/37483241
http://dx.doi.org/10.1021/acsomega.3c02387
work_keys_str_mv AT wangtao probabilisticbayesiandeeplearningapproachforonlineforecastingoffedbatchfermentation
AT youjiebing probabilisticbayesiandeeplearningapproachforonlineforecastingoffedbatchfermentation
AT gongxiugang probabilisticbayesiandeeplearningapproachforonlineforecastingoffedbatchfermentation
AT yangshanliang probabilisticbayesiandeeplearningapproachforonlineforecastingoffedbatchfermentation
AT wanglei probabilisticbayesiandeeplearningapproachforonlineforecastingoffedbatchfermentation
AT changzheng probabilisticbayesiandeeplearningapproachforonlineforecastingoffedbatchfermentation