KSBB Journal

The Korean Society for Biotechnology and Bioengineering (한국생물공학회)
권호 표지

Analysis of Two-Dimensional Fluorescence Spectra in Biotechnological Processes by Artificial Neural Networks II - Process Modeling using Backpropagation Neural Network -

인공신경망에 의만 생물공정에서 2차원 영광스펙트럼의 분석 II - 역전파 신경망에 의한 공정의 모델링 -

  • Lee Kum-Il (Department of Industrial Engineering, Chonnam National University, Research Center for Biophotonics) ;
  • Yim Yong-Sik (Department of Industrial Engineering, Chonnam National University, BioProcess Technology Lab.) ;
  • Sohn Ok-Jae (Department of Industrial Engineering, Chonnam National University, BioProcess Technology Lab.) ;
  • Chung Sang-Wook (Department of Industrial Engineering, Chonnam National University, Research Center for Biophotonics) ;
  • Rhee Jong Il (Faculty of Applied Chemical Engineering, Chonnam National University, BioProcess Technology Lab.)
  • 이금일 (전남대학교 산업공학과, 바이오광 기반기술개발 사업단) ;
  • 임용식 (전남대학교 물질 생물화공과, 생물공정기술연구실) ;
  • 손옥재 (전남대학교 물질 생물화공과, 생물공정기술연구실) ;
  • 정상욱 (전남대학교 산업공학과, 바이오광 기반기술개발 사업단) ;
  • 이종일 (전남대학교 응용화학공학부, 생물공정기술연구실)
  • Published : 2005.08.01
Download PDF
⟨ Previous Next ⟩

Abstract

A two-dimensional (2D) spectrofluorometer was used to monitor various fermentation processes with recombinant E. coli for the production of 5-aminolevulinic acid (ALA). The whole fluorescence spectral data obtained during a process were analyed using artificial neural networks, i.e. self-organizing map (SOM) and feedforward backpropagation neural network (BPNN).Based on the classified fluorescence spectra a supervised BPNN algorithm was used to predict some of the process parameters. It was also shown that the BPNN models could elucidate some sections of the process performance, e.g. forecasting the process performance.

본 연구에서는 인공신경망 알고리즘을 이용하여 생물공정에서 수집된 형광스펙트럼 데이터를 분류, 분석하고 공정변수들을 예측하기 위한 공정의 모델링에 대해서 논의하였다. SOM에 의해 분류된 전파장 스펙트럼 데이터들은 발효공정의 변수와 형광데이터 사이에 비선형관계를 설명하기 위하여 사용되었다. BPNN알고리즘은 SOM에서 분류된 데이터들을 입력자료로 이용하여 공정에 대한 모델식을 세우고, 이를 이용하여 배출가스 내 $CO_2$ 농도 및 발효액 중 세포농도와 같은 공정변수들을 예측하는데 사용되었다. 또한 BPNN 모델은 강력하면서도 훈련데이터의 범위를 넘어서는 공정의 데이터들을 예측할 수 있기 때문에 폭넓은 활용가능성을 가지고 있다.

Keywords

References

  1. Alaradi A. A., S. Rohani (2002), Identification and control of a riser-type FCC using neural networks, Com. & Chem. Eng. 26, 401-421 https://doi.org/10.1016/S0098-1354(01)00756-6
  2. Kobayashi H., R. Ozawa (2003), Adaptive neural network control of tendon-driven mechanisms with elastic tendons, Autornatica 39, 1509-1519
  3. Gontarski C. A., P. R. Rodrigues, M. Mori, L. F. Prenem (2000), Simulation of an industrial wastewater treatment plant using artificial neural networks, Com. & Chem. Eng. 24, 1719-1723 https://doi.org/10.1016/S0098-1354(00)00449-X
  4. Chen L. Z., S. K. Nguang, X. D. Chen, X. M. Li (2005), Modeling and optimization of fed-batch fermentation processes using dynamic neural networks and genetic algorithms, Biochem. Eng. J., in print
  5. Wolf G., J. S. Almeida, C. Pinheiro, V. Correia, C. Rodrigues, M. A. M. Reis, J. G. Crespo (2001), Two-dimensional fluorometry coupled with artificial neural networks: a novel method for on-line monitoring of complex biological processes, Biotechnol. Bioeng. 72, 297-306 https://doi.org/10.1002/1097-0290(20010205)72:3<297::AID-BIT6>3.0.CO;2-B
  6. Maltlab manual, vers. 6.1, The Mathworks, Inc., USA (2002)
  7. Lee K. I., Y. S. Yim, Kim C. K., S. H. Lee, S. W. Chung, J. I. Rhee (2005), Analysis of tow-dimensional fluorescence spectra in biotechnological ptocesses by artificial neural networks. 1. CaIssification of fluorescence spectra using self-organizing maps, Korean J. Biotechnol. Bioeng., submitted
  8. Baughman D. R., Y. A. Liu (1995), Neural Networks in bioprocessing and chemical engineering, Academic Press, San Diego, USA
  9. Rhee J. I., K. I. Lee, C. K. Kim, Y. S. Yim, S. W. Chung, J. Wei, K. H. Bellgardt (2005), Classification of two-dimensional fluorescence spectra using self-organizing maps, Biochem. Eng. J. 22, 135-144 https://doi.org/10.1016/j.bej.2004.09.008
  10. Tambe S. S., B. D. Kulkarni (1996), Elements of artificial neural networks with selected applications in chemical engineering, chemical & biological sciences, Simulation & Advanced Controls, Inc. Lousville, USA
  11. Basheer I. A., M. Hajmeer (2000), Artificial neural networks: fundamentals, computing, design, and application, J. Microbiol. Methods 43, 3-31 https://doi.org/10.1016/S0167-7012(00)00201-3
  12. Bhat N. V., T. J. McAvoy (1992), Determining model structure for neural network stripping, Com. & Chem. Eng. 16, 271-281 https://doi.org/10.1016/0098-1354(92)80047-D
  13. Servercan M., F. Severcan, P. I. Haris (2001), Estimation of protein secondary structure from FTIR spectra using neural networks, J. Molecul. Structure 565, 383-387 https://doi.org/10.1016/S0022-2860(01)00505-1
  14. Chung S. Y., K. H. Sea, J. I. Rhee (2005), Influence of culture conditions on the ptoduction of extra-cellular 5-aminolevulinic acid (ALA) by recombinant E. coli, Process Biochem. 40, 385-394 https://doi.org/10.1016/j.procbio.2004.01.024
  15. Nicholson D. J., D. 8. Kell, C. L. Davey (1996), Deconvolution of the dielectric spectra of microbial cell suspensions using multivariate calibration and artificial neural networks, Bioelectrochern. Bioener. 39, 185-193 https://doi.org/10.1016/0302-4598(95)01890-5
  16. Mukherjee J., C. Lindermann, T. Scheper (1999), Fluorescence monitoring during cultivation of Enterobacter aerogenes at different oxygen levels, Appl. Microbiol. Biotechnol. 52, 489-494 https://doi.org/10.1007/s002530051550