Abstract
The application of artificial neural networks to the estimation and prediction of bioprocess variables is presented in this paper. A neural network methodology is discussed, which uses environmental and physiological information available from on-line sensors, to estimate concentration of species in the bioreactor. Two case studies are presented, both based on the ethanol production by Zymomonas mobilis. An efficient optimization algorithm which reduces the number of iterations required for convergence is proposed. Results are presented for different training sets and different training methodologies. It is shown that the neural network estimator provides good on-line bioprocess state estimations.
Preview
Unable to display preview. Download preview PDF.
Abbreviations
- a:
-
Power of the ethanol inhibition term in μ
- b:
-
Power of the ethanol inhibition term in q p
- D h−1:
-
Dilution rate
- F l h−1:
-
Permeate flow
- F0 l h−1:
-
Total feed flow to the system
- Ki g l−1:
-
Substrate inhibition constant for growth
- K′i g l−1:
-
Substrate inhibition constant for ethanol production
- Ks g l−1:
-
Monod kinetic constant
- K′s g l−1:
-
Saturation constant for q p
- p g l−1:
-
Ethanol concentration
- P′i g l−1:
-
Ethanol threshold concentration for ethanol production
- Pm g l−1:
-
Maximum ethanol concentration for cell growth
- P′m g l−1:
-
Maximum ethanol concentration for ethanol production
- qp g g−1 h−1:
-
Specific ethanol production rate
- qpm g g−1 h−1:
-
Maximum specific ethanol production rate
- R:
-
Recycle ratio
- s g l−1:
-
Glucose concentration
- Si g l−1:
-
Threshold substrate concentration for cell growth
- s0 g l−1:
-
Glucose concentration in the feed stream
- x g l−1:
-
Biomass concentration
- xmax g l−1:
-
Maximum cell concentration
- Yp/s g g−1:
-
Ethanol yield, g ethanol per g substrate consumed
- μ h−1:
-
Specific growth rate
- μmo h−1:
-
Maximum specific growth rate at zero ethanol concentrations
- d p :
-
L-dimensional target vector
- E p :
-
Sum-of-squares error for training example p
- E(w):
-
Total sum-of-squares error for all input patterns, function of weight vector w
- G (q) :
-
Vector of steepest descent directions in iteration (q)
- g (q) p :
-
Gradient vector of one input-output pattern p
- g (q) uv :
-
Element of the gradient vector g (q)p
- n:
-
Number of interconnection weights in the network
- O pj :
-
Output of neuron j from the training set p
- p:
-
number of training pattern
- q:
-
number of iteration
- r (q) :
-
Vector of conjugate gradient directions in Eq. (17)
- S pj :
-
Activation state of neuron j from the training set p
- S (q) :
-
Vector of search directions for conjugate gradient algorithm
- w (q) :
-
Vector of neural network weights in iteration q
- w ij :
-
Interconnection weight from node i to node j
- x p :
-
N-dimensional network input vector
- y p :
-
L-dimensional network output vector
- z (q) :
-
Vector of conjugate gradient directions in Eq. (16)
- α(q) :
-
Step size in iteration q used in Eq. (8)
- β(q) :
-
Step size in iteration q used in Eq. (17)
- γ(q) :
-
Step size in iteration q used in Eq. (16)
- δ pv :
-
Change in E p due to changes in neural activation state S pu.
- σ:
-
Argument of the sigmoid function.
References
Ungar LH (1990) A bioreactor benchmark for adaptive network-based process control. In: Sutton Miller RS, Werbos WT, Werbos PJ (eds) Neural Networks for Control. MIT Press, 387
Ungar LH, Powell BA, Kamens SN (1990) Computers in Chemical Engineering 14:561
Venkatasubramanian V, Vaidyanathan R, Yamamoto Y (1990) Computers in Chemical Engineering 14: 699
Vaidyanathan R, Venkatasubramanian V (1990) Representing and diagnosing process trends using neural networks. In AIChE Annual Meeting, Chicago, Il, November 1990
Bhat N, McAvoy TJ (1990) Computers in Chemical Engineering, 14: 573
Rumelhart DE, Hinton GE, Williams RJ (1986) A general framework for parallel distributed processing. In: McClelland JL, Rumelhart DE, the PDP Research Group (eds) Parallel Distributed Processing: Explorations in the Micro structure of Cognition. MIT Press p 45
Rogers PL, Lee KJ, Skotnicki ML, Tribe DE (1982) Advances in Biochemical Engineering 23: 37
Carleysmith SW, Fox RI (1984) Advances in Biotechnological Processes 3: 1
Tenorio MF (1990) IEEE Transactions on Neural Networks 1: 100
Wang HY., Cooney CL, Wang DIC (1979) Biotechnology and Bioengineering 21: 975
Zabriskie DW, Humphrey AE (1978) AIChE Journal 24: 138
Wang NS, Stephanopoulos GN (1984) CRC Critical Reviews in Biotechnology 2: 1
Rolf MJ, Lim HC (1985) Biotechnology and Bioengineering 27: 1236
Andersen MY, Jorgensen SB (1989) Identification of a simulated continuous yeast fermentation. In: Fox RI, Fish NM, Thornhill NF (eds) Computer Applications in Fermentation Technology: Modelling and Control of Biotechnological Processes. Elsevier, Amsterdam 205
Petersen JN, Whyatt GA (1990) Biotechnology and Bioengineering 35: 712
Stephanopoulos G, San K-Y (1984) Biotechnology and Bioengineering 26: 1176
Montgomery PA, Williams D (1989) On-line estimation of cell mass using an Extended Kalman Filter. In: Fox RI, Fish NM, Thornhill NF (eds) Computer Applications in Fermentation Technology: Modelling and Control of Biotechnological Processes. Elsevier, Amsterdam, 221
Shimizu H, Takamatsu T, Shioya S, Suga K-I (1989) Biotechnology and Bioengineering 33: 354
Ryoo D, Murphy VG, Karim MN, Tengerdy RP (1991) Biotechnology and Bioengineering. In press.
Bastin G, Dochain D (1986) Automatica 22: 705
Pomerleau Y, Perrier M (1989) Nonlinear estimation and adaptive control of a fed-batch fermentor. In: Fox RI, Fish NM, Thornhill NF (eds) Computer Applications in Fermentation Technology: Modelling and Control of Biotechnological Processes. Elsevier, Amsterdam, 361
Qian N, Sejnowski TJ (1988) Journal of Molecular Biology 202: 865
McAvoy TJ, Wang NS, Naidu S, Bhat N, Gunter J, Simmons M (1989) Interpreting biosensor data via backpropagation. In: IJCNN International Joint Conference on Neural Networks, San Diego, CA, vol 1 p 227
Hoskins JC, Himmelblau DM (1988) Computers in Chemical Engineering 12: 881
Haesloop D (1990) A neural network structure for system identification. In: Proceedings of the 1990 American Control Conference, May 23–25 p 2460
Breusegem VV, Thibault J, Cheruy A (1991) Canadian Journal of Chemical Engineering 69: 481
Thibault J, Breusegem VV, Cheruy A (1990) Biotechnology and Bioengineering 36: 1041
MacGregor JF, Marlin TE, Kresta JV (1991) Some comments of neural networks and other empirical modelling methods. In: Arkun Y, Ray WH (eds) Chemical Process Control CPC-IV, CACHE and AIChE, New York, p 665
Leonard J, Kramer MA (1990) Computers in Chemical Engineering 14: 337
Le D (1985) Mathematical Programming 32: 41
Rivera SL, Karim MN (1990) Application of dynamic programming for fermentative ethanol production by Zymomonas mobilis. In: Proceedings of the 1990 American Control Conference, San Diego, CA, May 23–25, p 2144
Swings J, De Ley J (1977) Bacteriological Reviews 41: 1
Cybenko G (1989) Mathematics of Control, Signal and Systems 2: 303
Hertz J, Krogh A, Palmer RG (1991) Introduction to the Theory of Neural Computation, vol. I. Addison-Wesley, Redwood City, CA
Morris AJ, Montague GA, Tham MT, Aynsley M, Di Massimo C, Lant P (1991) Towards improved process supervision-algorithms and knowledge based systems. In: Arkun Y, Ray WH (eds) Chemical Process Control CPC-IV, CACHE and AIChE, New York, p 585
Lee KJ, Rogers PL (1983) The Chemical Engineering Journal 27: B31
Jeong JW, Snay J, Ataai MM (1990) Biotechnology and Bioengineering 35: 160
Veeramallu U, Agrawal P (1990) Biotechnology and Bioengineering 36: 694
Fahlman SE, Lebiere C (1990) The cascade-correlation learning architecture. In: Touretzky D (ed) Advances in Neural Information Processing Systems 2, Morgan Kauffman, San Mateo, p 524
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1992 Springer-Verlag
About this chapter
Cite this chapter
Karim, M.N., Rivera, S.L. (1992). Artificial neural networks in bioprocess state estimation. In: Modern Biochemical Engineering. Advances in Biochemical Engineering/Biotechnology, vol 46. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0000703
Download citation
DOI: https://doi.org/10.1007/BFb0000703
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-55276-5
Online ISBN: 978-3-540-47005-2
eBook Packages: Springer Book Archive