Phase Identification–Based Prediction of Product Formation for 2-Keto-l-Gulonic Acid Fermentation Processes

  • Lei CuiEmail author
  • Tao Xu
  • Zhifeng Wang
Original Research Paper


Because of the time-varying characteristics of 2-keto-l-gulonic acid (2-KGA) fermentation processes, phase identification–based prediction of product formation is presented in this paper. Firstly, based on the online measurement data of fermentation processes, fermentation stage is identified with the modified principal component analysis method. The state variable such as the product formation is predicted using the support vector machine (SVM) approach. The obtained stage information is fed back to the state variable prediction model. In different fermentation phases, the training databases and parameters of SVM for the prediction model are applied differently. With the data from industrial 2-KGA fermentation processes, pseudo-online prediction is practiced. The results indicate that the prediction approach has good performance.


Industrial fermentation process Principal component analysis Phase identification State variable prediction 


Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.


  1. Aiguo J, Peiji G (1998) Synthesis of 2-keto-L-gulonic acid from gluconic acid by co-immobilized Gluconobacter oxydans and Corynebacterium sp. Biotechnol Lett 20(10):939–942CrossRefGoogle Scholar
  2. Bremus C, Herrmann U, Bringer-Meyer S, Sahm H (2006) The use of microorganisms in L-ascorbic acid production. J Biotechnol 124(1):196–205CrossRefGoogle Scholar
  3. Browne MW (2000) Cross-validation methods. J Math Psychol 44(1):108–132MathSciNetCrossRefGoogle Scholar
  4. Cedeno MV, Rodriguez Aguilar LPF, Sanchez MC (2016) Bioprocess statistical control: identification stage based on hierarchical clustering. Process Biochem 51(12):1919–1929CrossRefGoogle Scholar
  5. Ge Z, Song Z, Gao F (2013) Review of recent research on data-based process monitoring. Ind Eng Chem Res 52(10):3543–3562CrossRefGoogle Scholar
  6. Hancock RD, Viola R (2002) Biotechnological approaches for L-ascorbic acid production. Trends Biotechnol 20(7):299–305CrossRefGoogle Scholar
  7. León-Roque N, Abderrahim M, Nuñez-Alejos L, Arribas SM, Condezo-Hoyos L (2016) Prediction of fermentation index of cocoa beans (Theobroma cacao L.) based on color measurement and artificial neural networks. Talanta 161:31–39CrossRefGoogle Scholar
  8. Liu G, Zhou D, Xu H, Mei C (2010) Model optimization of SVM for a fermentation soft sensor. Expert Syst Appl 37(4):2708–2713CrossRefGoogle Scholar
  9. Luttmann R, Bracewell DG, Cornelissen G, Gernaey KV, Glassey J, Hass VC, Kaiser C, Preusse C, Striedner G, Mandenius CF (2012) Soft sensors in bioprocessing: a status report and recommendations. Biotechnol J 7(8):1040–1048CrossRefGoogle Scholar
  10. Ning WZ, Tao ZX, Wang CH, Wang SD, Yan ZZ, Yin GL (1988) Fermentation process for producing 2-keto-L-gulonic acid. European Patent 0278447Google Scholar
  11. Rosales-Colunga LM, García RG, Rodríguez ADL (2010) Estimation of hydrogen production in genetically modified E. coli fermentations using an artificial neural network. Int J Hydrog Energy 35(24):13186–13192CrossRefGoogle Scholar
  12. Wang X, Chen J, Liu C, Pan F (2010) Hybrid modeling of penicillin fermentation process based on least square support vector machine. Chem Eng Res Des 88(4):415–420CrossRefGoogle Scholar
  13. Yao Y, Gao F (2009) A survey on multistage/multiphase statistical modeling methods for batch processes. Annu Rev Control 33(2):172–183CrossRefGoogle Scholar
  14. Yin S, Li X, Gao H (2015) Data-based techniques focused on modern industry: an overview. IEEE Trans Ind Electron 62(1):657–667CrossRefGoogle Scholar
  15. Yuan JQ, Vanrolleghem PA (1999) Rolling learning-prediction of product formation in bioprocesses. J Biotechnol 69(1):47–62CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.College of EngineeringShanghai Polytechnic UniversityShanghaiChina

Personalised recommendations