On-line fault diagnosis for optimal rice α-amylase production process of a temperature-sensitive mutant of Saccharomyces cerevisiae by an autoassociative neural network

Shimizu, Hiroshi; Yasuoka, K.; Uchiyama, Kimio; Shioya, Suteaki

doi:10.1016/s0922-338x(97)82997-2

Cited by 24 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus p corresponds to the total sampling time of the fermentation. The dimension of the mapping space, m , is set to 2 because the total variance explained by the first and second principal components was higher than 95% (Ignova et al, 1995; Shimizu et al, 1997).…”

Section: Methodsmentioning

confidence: 99%

Bioprocess Fault Detection by Nonlinear Multivariate Analysis: Application of an Artificial Autoassociative Neural Network and Wavelet Filter Bank

et al. 1998

View full text Add to dashboard Cite

A nonlinear multivariate analysis, artificial autoassociative neural network (AANN), was applied to bioprocess fault detection. In an optimal production process of a recombinant yeast with a temperature controllable expression system, faults in test cases with faulty temperature sensors and plasmid instability of recombinant cells could be detected by the AANN. Since the raw data of measured variables included high-frequency noise, a wavelet filter bank (WFB) was applied to noise elimination before training of the AANN. The filtering performance of the WFB was compared with those of some classical first-order digital filters. The filtered signals at several resolution scales by the WFB were employed as the training data of the AANN. The computing time and summation of square of errors in training were compared, and the appropriate degree of the noise filtering and the density of the training data of the AANN were discussed. The performance of the feature capturing by the AANN was compared with that by a linear multivariate analysis, principal component analysis. A J index defined in this paper, using inputs and outputs of the AANN, was used for fault detection successfully. The output of the first unit of the trained AANN functioned effectively for the discrimination of the data in the abnormal cases from the data in the normal cases.

show abstract

Section: Methodsmentioning

confidence: 99%

Bioprocess Fault Detection by Nonlinear Multivariate Analysis: Application of an Artificial Autoassociative Neural Network and Wavelet Filter Bank

et al. 1998

View full text Add to dashboard Cite

show abstract

“…It was concluded that the hybrid approach gives a better representation of the fermentation than the Elman neural network. Several authors have used neural networks as a fault detection tool (Ignova et al, 1997;Shimizu et al, 1997;Shimizu et al, 1998).…”

Section: Applications To Bioprocessesmentioning

confidence: 99%

Data‐Based Modeling and Analysis of Bioprocesses: Some Real Experiences

Karim

Hodge

Simon

2003

Biotechnology Progress

View full text Add to dashboard Cite

Data-generated models find numerous applications in areas where the speed of collection and logging of data surpasses the ability to analyze it. This work is meant to addresses some of the challenges and difficulties encountered in the practical application of these methods in an industrial setting and, more specifically, in the bioprocess industry. Neural network and principal component models are the two topics that are covered in detail in this paper. A review of these modeling technologies as applied to bioprocessing is provided, and four original case studies using industrial fermentation data are presented that utilize these models in the context of prediction and monitoring of bioprocess performance.

show abstract

“…Multi-variables clustering and analysis technique has been adopted in fault alarming/diagnosis and sensitive physiological variables selection in a couple of bioprocesses [15][16][17][18][19], aiming to better monitor process performance, optimize process control strategy, and detect operation fault to achieve stable production in maximal available extent. The cores of those multi-variables clustering methods basically lie on the followings: (1) classification of ''normal'' and ''abnormal'' fermentation runs based on certain performance index (i.e., yield, concentration) and judgment criteria; (2) construction of the reference database/model using the known ''normal fermentation data'' set and with helps of either PCA (i.e., KPCA, Kernel FDA) [15,16] or auto-associative neural network [17][18][19]; implementation of the compressed data comparison in between a new fermentation runs and the reference database so as to detect operational fault as early as possible.…”

Section: Introductionmentioning

confidence: 99%

“…The cores of those multi-variables clustering methods basically lie on the followings: (1) classification of ''normal'' and ''abnormal'' fermentation runs based on certain performance index (i.e., yield, concentration) and judgment criteria; (2) construction of the reference database/model using the known ''normal fermentation data'' set and with helps of either PCA (i.e., KPCA, Kernel FDA) [15,16] or auto-associative neural network [17][18][19]; implementation of the compressed data comparison in between a new fermentation runs and the reference database so as to detect operational fault as early as possible. The major drawbacks of the above-mentioned methods could be summarized as: (1) direct utilization difficulty due to their complicated algorithm; (2) difficulty in interpreting the biological mechanisms due to their black or grey box's nature; (3) classification failure in some cases if without certain special empiricism-based data pretreatment [18].…”

Section: Introductionmentioning

confidence: 99%

Effective and stable porcine interferon-α production by Pichia pastoris fed-batch cultivation with multi-variables clustering and analysis

Dong

Zhu

et al. 2009

Bioprocess Biosyst Eng

View full text Add to dashboard Cite

Efficient porcine interferon-alpha (pIFN-alpha) expression in high density recombinant Pichia pastoris cultivation was achieved in a 5 l bench-scaled bioreactor. The results indicated that a high and stable oxygen uptake rate (OUR) during induction phase was closely related with pIFN-alpha production efficiency. The multi-variables clustering and analysis results showed that the achievement of a high and stable OUR relied on a higher glycerol consumption rate during fed-batch culture phase and a moderate methanol level (around 10 g/l) during induction phase. In the high and stable OUR environments (200-300 mmol/l/h), the highest pIFN-alpha antiviral activity could reach a level of 6.7 x 10(6) IU/ml, which was more than 10-300-folds higher than those obtained at lower OUR (80-200 mmol/l/h) using the same bioreactor and those obtained in shaking flasks. Clustering and analysis of the specific growth and glycerol consumption rates data during culture phase could detect the ill fermentation state at early stage, potentially provided a simple and effective fault alarming/diagnosis method for the achievement of stable pIFN-alpha production.

show abstract

On-line fault diagnosis for optimal rice α-amylase production process of a temperature-sensitive mutant of Saccharomyces cerevisiae by an autoassociative neural network

Cited by 24 publications

References 12 publications

Bioprocess Fault Detection by Nonlinear Multivariate Analysis: Application of an Artificial Autoassociative Neural Network and Wavelet Filter Bank

Bioprocess Fault Detection by Nonlinear Multivariate Analysis: Application of an Artificial Autoassociative Neural Network and Wavelet Filter Bank

Data‐Based Modeling and Analysis of Bioprocesses: Some Real Experiences

Effective and stable porcine interferon-α production by Pichia pastoris fed-batch cultivation with multi-variables clustering and analysis

Contact Info

Product

Resources

About