Soft Sensors Development For On-line Bioreactor State Estimation

During a fermentation process, variables such as concentrations are determined by off-line laboratory analysis, making this set of variables of limited use for control purposes. However, these variables can be on-line estimated using soft sensors. The objective of this study is to present the state of the art of state estimator techniques. Special attention was given to filtering techniques, namely extended Kalman filter, adaptive observers, and artificial neural networks (ANN). It is shown that software based state estimation is a powerful technique that can be successfully used to enhance automatic control performance of biological systems as well as in system monitoring and on-line optimisation. (C) 2000 Elsevier Science Ltd.During a fermentation process, variables such as concentrations are determined by off-line laboratory analysis, making this set of variables of limited use for control purposes. However, these variables can be on-line estimated using soft sensors. The objective of this study is to present the state of the art of state estimator techniques. Special attention was given to filtering techniques, namely extended Kalman filter, adaptive observers, and artificial neural networks (ANN). It is shown that software based state estimation is a powerful technique that can be successfully used to enhance automatic control performance of biological systems as well as in system monitoring and on-line optimisation.2402/07/1510991103Albiol, J., Robusté, J., Casas, C., Poch, M., Biomass estimation in plant cell cultures using an extended Kalman filter (1993) Biotechnology Progress, 9, pp. 174-178Albiol, J., Campmajó, C., Casas, C., Poch, M., Biomass estimation in plant cell cultures: A neural network approach (1995) Biotechnology Progress, 11, pp. 88-92Bastin, G., Dochain, D., (1990) On-line Estimation and Adaptive Control of Bioreactors, , Amsterdam: ElsevierBellgardt, K.H., Kuhlmann, W., Meyer, H.D., Schügerl, K., Thoma, M., Application of an extended Kalman filter for state estimation of a yeast fermentation (1986) IEE Proceedings, 133 (5), pp. 226-234Breusegem, V., Thibault, J., Chéruy, A., Adaptive neural models for on-line prediction in fermentation (1991) The Canadian Journal of Chemical Engineering, 69, pp. 481-487Chéruy, A., Software sensors in bioprocess engineering (1997) Journal of Biotechnology, 52, pp. 193-199Di Massimo, C., Montague, G.A., Willis, M.J., Tham, M.T., Morris, A.J., Towards improved penicillin fermentation via artificial neural networks (1992) Computers & Chemical Engineering, 16 (4), pp. 283-291Fu, P.-C., Barford, J.P., A hybrid neural network - First principles approach for modelling of cell metabolism (1996) Computers & Chemical Engineering, 20 (6-7), pp. 951-958Gee, D.A., Ramirez, W.F., On-line state estimation and parameter identification for batch fermentation (1996) Biotechnology Progress, 12, pp. 132-140Glassey, J., Ignova, M., Ward, A.C., Montague, G.A., Morris, A.J., Bioprocess supervision: Neural networks and knowledge-based, systems (1997) Journal of Biotechnology, 52, pp. 201-205Gudi, R.D., Shah, S., Gray, M., Adaptive multirate state and parameter estimation strategies with application to a bioreactor (1995) American Institute of Chemical Engineering Journal, 41 (11), pp. 2451-2464Jazwinski, A.H., (1970) Stochastic Processes and Filtering Theory, , New York, USA: Academic PressKalman, R.E., Bucy, R.S., New results in linear filtering and prediction theory (1961) Transactions ASME: Journal of Basic Engineering, 83, pp. 95-108Karim, M.N., Rivera, S.L., Comparison of feed-forward and recurrent neural networks for bioprocess state estimation (1992) Computers & Chemical Engineering, 16 (SUPPL.), pp. S369-S377McKay, B., Willis, M., Montague, G., Barton, B., Using genetic programming to develop inferential estimation algorithms (1996) GP'96, , San Francisco, USAMontague, G.A., Morris, A.J., Wright, A.R., Aynsley, M., Ward, A.C., Online estimation and adaptive control of penicillin fermentation (1986) IEE Proceedings, 133 (5), pp. 240-246Montague, G.A., Morris, A.J., Tham, M.T., Enhancing bioprocess operability with generic software sensors (1992) Journal of Biotechnology, 25, pp. 183-201Pons, M.N., Rajab, A., Flaus, J.M., Engasser, J.M., Cheruy, A., Comparison of estimation methods for biotechnological processes (1988) Chemical Engineering Science, 43 (8), pp. 1909-1914San, K.-Y., Stephanopoulos, G., Studies on on-line bioreactor identification. II. Numerical and experimental results (1984) Biotechnology & Bioengineering, 26, pp. 1189-1197Shimizu, K., A tutorial review on bioprocess systems engineering (1996) Computers & Chemical Engineering, 20 (6-7), pp. 915-941Shimizu, H., Takamatsu, T., Shioya, S., Suga, K.-I., An algorithmic approach to constructing the on-line estimation system for the specific growth rate (1989) Biotechnology & Bioengineering, 33, pp. 354-364Stephanopoulos, G., San, K.-Y., Studies on on-line bioreactor identification. I. Theory (1984) Biotechnology & Bioengineering, 26, pp. 1176-1188Syu, M.-J., Tsao, G.T., Neural network modeling of batch cell growth pattern (1993) Biotechnology & Bioengineering., 42, pp. 376-380Tham, M.T., Morris, A.J., Montague, G.A., Soft-sensing: A solution to the problem of measurement delays (1989) Chemical Engineering Research & Design, 67, pp. 547-554Thibault, J., Breusegem, V.V., Chéruy, A., On-line prediction of fermentation variables using neural networks (1990) Biotechnology & Bioengineering, 36, pp. 1041-1048Tholudur, A., Ramirez, W.F., Optimisation of fed-batch bioreactors using neural network parameter function models (1996) Biotechnology Progress, 12, pp. 302-309Yet-Pole, I., Wu, W.-T., Liu, Y.-C., Neural network modelling for on-line state estimation in fed-batch culture of L-lysine production (1996) The Chemical Engineering Journal, 61, pp. 35-40Zhao, Y., (1996) Studies on Modelling and Control of Continuous Biotechnical Processes, , Ph.D. thesis, Norwegian University of Science and Technology, NorwayZorzetto, L.F.M., (1995) Bioprocess Monitoring with Hybrid Neural Network/mechanistic Model Based State Estimators, , Ph.D. thesis. University of Nottingham, UKZorzetto, L.F.M., Maciel Filho, R., Modelling a batch beer production system, through artificial neural networks and hybrid models, using distinct levels of information (1997) The First European Congress on Chemical Engineering, pp. 2535-2538. , ECCE-1, Florence, Ital