Skip to main content
SpringerLink
Log in

Fault diagnosis of a benchmark fermentation process: a comparative study of feature extraction and classification techniques

  • Original Paper
  • Published:
Bioprocess and Biosystems Engineering Aims and scope Submit manuscript

Abstract

This paper investigates fault diagnosis in batch processes and presents a comparative study of feature extraction and classification techniques applied to a specific biotechnological case study: the fermentation process model by Birol et al. (Comput Chem Eng 26:1553–1565, 2002), which is a benchmark for advanced batch processes monitoring, diagnosis and control. Fault diagnosis is achieved using four approaches on four different process scenarios based on the different levels of noise so as to evaluate their effects on the performance. Each approach combines a feature extraction method, either multi-way principal component analysis (MPCA) or multi-way independent component analysis (MICA), with a classification method, either artificial neural network (ANN) or support vector machines (SVM). The performance obtained by the different approaches is assessed and discussed for a set of simulated faults under different scenarios. One of the faults (a loss in mixing power) could not be detected due to the minimal effect of mixing on the simulated data. The remaining faults could be easily diagnosed and the subsequent discussion provides practical insight into the selection and use of the available techniques to specific applications. Irrespective of the classification algorithm, MPCA renders better results than MICA, hence the diagnosis performance proves to be more sensitive to the selection of the feature extraction technique.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14

Similar content being viewed by others

Abbreviations

A :

Unknown mixing matrix

C a, C b :

Acid and Base molarity

\( C_{{{\text{CO}}_{2} }} \) :

Carbon dioxide concentration

C H :

Hydrogen ion concentration

C L :

Dissolved oxygen concentration

dt :

Derivative time in the control system

E :

Residuals

E N :

Error signal at sample N

F :

Feed flow rate of substrate

F a :

Acid flow rate

F b :

Base flow rate

F c :

Heating/cooling water flow rate

f g :

Air flow rate

F1:

Diagnosis performance index

f :

Transfer function in the ANN

I :

Number of batches

I k :

Integral signal at sample k

I k−1 :

Previous integral signal

J :

Process variables

k :

Neuron

K :

Time observations

K c :

Proportional gain in PI control

K x :

Contois saturation constant

p :

Loading vectors in PCA

P :

Penicillin concentration

P k :

Proportional signal at sample k

P w :

Power density

Q i :

Sum of the squared errors or Q statistic

Q rxn :

Reaction heat rate

Q α :

Control limit of the Q statistic

r :

Total number of inputs to the neuron

R :

Number of principal and independent components and

S :

Substrate concentration, covariance matrix and independent component matrix

s f :

Feed substrate concentration

T :

Reactor temperature

t :

Scores

T f :

Feed temperature of substrate

T 2 i :

Hotelling’s T squared for each batch

T 2lim :

T squared control limit

U :

Control signal

V :

Culture volume

v k :

Input to the transfer function of neuron

w kj :

Input weights to neuron

X :

Biomass concentration and unfolded data matrix

X * :

Centered and scaled data matrix

\( \bar{X} \) :

Three-dimensional data matrix

x j :

Output values from the previous layer in the ANN

Y k :

Current value of the controlled variable at the sampling time

Y SP :

Set point of the controlled variable

λ i :

Eigenvalues of the covariance matrix

τI :

Integral constant in the PI control

References

  1. Birol G, Ündey C, Çinar A (2002) A modular simulation package for fed-batch fermentation: penicillin production. Comput Chem Eng 26:1553–1565

    Article  CAS  Google Scholar 

  2. Yoo C, Lee P, Vanrolleghem PA, Lee I (2004) On-line monitoring of batch processes using multiway independent component analysis. Chemom Intell Lab Syst 71:151–163

    Article  CAS  Google Scholar 

  3. Tian X, Zhang X, Deng X, Chen S (2009) Multiway kernel independent component analysis based on feature samples for batch process monitoring. Neurocomputing 72:1584–1596

    Article  Google Scholar 

  4. Nucci ER, Cruz AJG, Giordano RC (2010) Monitoring bioreactors using principal component analysis: production of penicillin G acylase as a case study. Bioprocess Biosyst Eng 33:557–564

    Article  CAS  Google Scholar 

  5. Pau L (1981) Failure diagnosis and performance monitoring. Marcel Dekker, New York

    Google Scholar 

  6. Himmelblau DM (1978) Fault detection and diagnosis in chemical and petrochemical processes. Elsevier, Amsterdam

    Google Scholar 

  7. Kourti T (2002) Process analysis and abnormal situation detection: from theory to practice. IEEE Control Syst Mag 22:10–25

    Article  Google Scholar 

  8. Cinar A, Parulekar S, Undey C, Birol G (2003) Batch fermentation: modeling, monitoring and control. Marcel Dekker, New York

    Google Scholar 

  9. Westerhuis J, Kourti T, MacGregor J (1999) Comparing alternative approaches for multivariate statistical analysis of batch process data. J Chemom 13:379–396

    Article  Google Scholar 

  10. Dahl S, Piovoso M, Kosanovich K (1999) Translating third-order data analysis methods to chemical batch processes. Chemom Intell Syst 46:161–180

    Article  CAS  Google Scholar 

  11. Nomikos P, MacGregor J (1995) Multivariate SPC charts for monitoring batch processes. Technometrics 37:41–59

    Google Scholar 

  12. Nomikos P, MacGregor J (1994) Monitoring of batch processes using multiway principal component analysis. AIChE J 40:1361–1375

    Article  CAS  Google Scholar 

  13. Neogi D, Schlags CE (1998) Multivariate statistical analysis of an emulsion batch process. Ind Eng Chem Res 37:3971–4979

    Article  CAS  Google Scholar 

  14. Tates AA, Louwerse DJ, Smilde AK, Koot GLM, Berndt H (1999) Monitoring a PVC batch process with multivariate statistical process control charts. Ind Eng Chem Res 38:4769–4776

    Article  CAS  Google Scholar 

  15. Van Sprang E, Ramaker H, Westerhuis J, Gurden S, Smilde A (2002) Critical evaluation of approaches for on-line batch process monitoring. Chem Eng Sci 57:3979–3991

    Article  Google Scholar 

  16. Ündey C, Tatara E, Çinar A (2003) Real-time batch process supervision by integrated knowledge-based systems and multivariate statistical methods. Eng Appl Artif Intell 16:555–566

    Article  Google Scholar 

  17. Venkatasubramanian V, Rengaswamy R, Yin K, Kavuri SN (2003) A review of process fault detection and diagnosis. Part I: quantitative model-based methods. Comput Chem Eng 27:293–311

    Article  CAS  Google Scholar 

  18. Pokkinen M, Flores Z, Asama H, Endo I, Aarts R, Linko P (1992) A knowledge based system for diagnosing microbial activities during a fermentation process. Bioprocess Biosyst Eng 7:331–334

    CAS  Google Scholar 

  19. Román RC, Hernández OG, Urtubia UA (2011) Prediction of problematic wine fermentations using artificial neural networks. Bioprocess Biosyst Eng 34:1057–1065

    Google Scholar 

  20. Geladi P, Isaksson H, Lindqvist L, Wold S, Esbensen K (1989) Principal component analysis of multivariate images. Chemom Intell Lab Syst 5:209–220

    Article  Google Scholar 

  21. MacGregor JF, Nomikos P (1991) Monitoring batch processes. In: Reklaitis GV et al (eds) Batch processing systems engineering, NATO ASI Series F, vol 143. pp 241–258

  22. Qin S (2003) Statistical process monitoring: basics and beyond. J Chemom 17:480–502

    Article  CAS  Google Scholar 

  23. Harington J (1975) Clustering algorithms. Wiley, New York

    Google Scholar 

  24. Aguado D, Ferrer J, Seco A (2007) Multivariate SPC of a sequencing batch reactor for wastewater treatment. Chemom Intell Lab Syst 85:82–93

    Article  CAS  Google Scholar 

  25. Villez K, Steppe K, De Pauw DJW (2009) Use of unfold PCA for on-line plant stress monitoring and sensor failure detection. Biosyst Eng 103:23–34

    Article  Google Scholar 

  26. Yoo C, Lee D, Vanrolleghem PA (2004) Application of multiway ICA for on-line process monitoring of a sequencing batch reactor. Water Res 38:1715–1732

    Article  CAS  Google Scholar 

  27. Venkatasubramanian V, Chan K (1989) A neural network methodology for process fault diagnosis. AIChE J 35:1993–2002

    Article  CAS  Google Scholar 

  28. Kavuri S, Venkatasubramanian V (1993) Representing bounded fault classes using neural networks with ellipsoidal activation functions. Comput Chem Eng 17:139–163

    Article  CAS  Google Scholar 

  29. Hoskins JC, Himmelblau DM (1988) Artificial neural network models of knowledge representation in chemical engineering. Comput Chem Eng 2:881–890

    Article  Google Scholar 

  30. Watanabe K, Matsuura I, Abe M, Kubota M, Himmelblau DM (1989) Incipient fault diagnosis of chemical processes via artificial neural networks. AIChE J 35:1803–1812

    Article  CAS  Google Scholar 

  31. Ruiz D, Nougués JM, Calderón Z, Espuña A, Puigjaner L (2000) Neural network based framework for fault diagnosis in batch chemical plants. Comput Chem Eng 24:777–784

    Article  CAS  Google Scholar 

  32. Haykin S (1994) Neural networks. A comprehensive foundation. Macmillan College Publishing, New York

    Google Scholar 

  33. Leger R, Garland Wm, Poehlman W (1998) Fault detection and diagnosis using statistical control charts and artificial neural networks. Artif Intell Eng 12:35–47

    Article  Google Scholar 

  34. Vapnik V (1999) The nature of statistical learning theory. Springer, New York

    Google Scholar 

  35. Vapnik V (1998) Statistical learning theory. Wiley, New York

    Google Scholar 

  36. Yélamos I, Graells M, Puigjaner L, Escudero G (2007) Simultaneous fault diagnosis in chemical plants using a multilabel approach. AIChE J 53:2871–2884

    Article  Google Scholar 

  37. Chiang L, Kotanchek M, Kordon A (2004) Fault diagnosis based on Fisher discriminant analysis and support vector machines. Comput Chem Eng 28:1389–1401

    Article  CAS  Google Scholar 

  38. Kulkarni A, Jayaraman V, Kulkarni B (2005) Knowledge incorporated support vector machines to detect faults in Tennessee Eastman Process. Comput Chem Eng 29:2128–2133

    Article  CAS  Google Scholar 

  39. Yunfeng L, Zhifeng W, Jingqi Y (2006) On-line fault detection using SVM-based dynamic MPLS for batch processes. Chin J Chem Eng 14:754–758

    Article  Google Scholar 

  40. Ramaker H, Van Sprang E, Gurden S, Westerhuis J, Smilde A (2002) Improved monitoring of batch processes by incorporating external information. J Process Control 12:569–576

    Article  CAS  Google Scholar 

  41. Jolliffe IT (1986) Principal component analysis. Springer, New York

    Google Scholar 

  42. MacGregor JF, Kourti T (1995) Statistical process control of multivariate processes. Control Eng Pract 3:403–414

    Article  Google Scholar 

  43. Jackson JE, Mudholkar GS (1979) Control procedures for residuals associated with principal component analysis. Technometrics 21:341–349

    Google Scholar 

  44. Yoo C, Villez K, Lee I, Rosén C (2007) Multi-model statistical process monitoring and diagnosis of a sequencing batch reactor. Biotechnol Bioeng 96:687–701

    Article  CAS  Google Scholar 

  45. Montgomery DC (2005) Introduction to statistical quality control, 5th edn. Wiley International, USA

    Google Scholar 

  46. Camacho J, Picó J, Ferrer A (2009) The best approaches in the on-line monitoring of batch processes based on PCA: does the modelling structure matter? Anal Chim Acta 642:59–68

    Article  CAS  Google Scholar 

  47. Jackson JE (1991) A user’s guide to principal components. Wiley, USA

    Book  Google Scholar 

  48. Kent A, Berry M, Luehrs F, Perry J (1955) Machine literature searching: VIII. Operational criteria for designing information retrieval systems. American Documentation 6:93–101

    Article  Google Scholar 

  49. Monroy I, Benitez R, Escudero G, Graells M (2010) A semi-supervised approach to fault diagnosis for chemical processes. Comput Chem Eng 34:631–642

    Article  CAS  Google Scholar 

  50. Huang S-C, Huang Y-F (1991) Bounds on the number of hidden neurons in multilayer perceptrons. IEEE Trans Neural Netw 2:47–55

    Article  CAS  Google Scholar 

Download references

Acknowledgments

Financial support from Generalitat de Catalunya through the FI fellowship program (2011F1_B200181) is fully appreciated. Financial support through the research projects TolerantT (DPI 2006-05673) and EHMAN (DPI2009-09386) funded by the European Union (European Regional Development Fund 2007-13) and the Spanish Ministry of Science and Innovation is fully appreciated. The MPCA Matlab code was developed at modelEAU, Université Laval, Québec to whom also acknowledge.

Author information

Authors and Affiliations

  1. Chemical Engineering Department (DEQ)-CEPIMA, Universitat Politècnica de Catalunya, EUETIB Comte d’Urgell 187, 08036, Barcelona, Spain

    Isaac Monroy & Moisès Graells

  2. Laboratory for Intelligent Process Systems, School of Chemical Engineering, Purdue University, 480 Stadium Mall Drive, West Lafayette, IN, 47907, USA

    Kris Villez & Venkat Venkatasubramanian

Authors
  1. Isaac Monroy
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Kris Villez
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Moisès Graells
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Venkat Venkatasubramanian
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Isaac Monroy.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Monroy, I., Villez, K., Graells, M. et al. Fault diagnosis of a benchmark fermentation process: a comparative study of feature extraction and classification techniques. Bioprocess Biosyst Eng 35, 689–704 (2012). https://doi.org/10.1007/s00449-011-0649-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00449-011-0649-1

Keywords

Search

Navigation