Multiparameter polynomial models for monitoring and management of fermentation and food processes

The project belongs to a larger Dynamic Systems Modeling (DYSMO) project coordinated by VTT Automation. DYSMO is one of the projects of Adaptive and Intelligent Systems Applications 1994-1999 -program of TEKES.

Project description

The primary goal was to develop model based solutions for monitoring and management of fermentation processes producing enzymes and for chromatographic separation processes. Both processes are difficult to model with traditional methods. These processes were selected to be used as testbeds for developing dynamic NN-models based on Volterra and Wiener theories for nonlinear dynamic systems.

The original Wiener representation consists of linear dynamics (Laguerre filters) and static nonlinear mapping (polynomial expansion). In Wiener-NN, static nonlinear mapping is approximated with NN. The feedforward Wiener-MLP is suitable for modeling of finite memory systems. The Wiener-MLP with feedback can also be used for modeling of autonomous type systems, like fermentation processes.

Due to robust orthogonal (e.g. Laguerre) description of signals, Wiener-MLP or -SOM classifiers are advantageous for recognizing of spatio-temporal patterns needed in process monitoring. A method using Wiener-type MLP or SOM classifiers for detecting and recognizing the functional states on-line was developed. It was demonstrated with Bacillus subtilis fermentation.

The Laguerre descriptions of the signals realize the state in different Wiener-NNs. In the NOE case, the state-space formulation was utilized in parameter estimation with the Extended Kalman filter, which can be also used directly as a state estimator in connection with the NFIR and NOE-type Wiener-NN models.

The various Wiener-models were applied to model the dynamics of industrial chromatographic separation columns and industrial Tricoderma fungi fermentation for monitoring, state estimation, on-line simulation and fault diagnosis. Chromatographic separation process was also modeled with hybrid models: some parameter dependencies in the traditional mechanistic model (partial differential equation) were described with different black-box mappings.

More information

Publications

Publication in series

• Arto Visala,
  Modeling of nonlinear processes using Wiener-NN representation and multiple models,
  Espoo: TKK/Automaatio, 1997. (Automation technology laboratory, Research reports, 18).
• A Visala,
  Open Communication in Manufacturing Automation,
  Espoo: Valtion teknillinen tutkimuskeskus, 1993. (VTT Research Notes, 1471).
• X Zhang, A Visala and A Halme,
  A Kinetic Model of Mammalian Cell Cultures and Fitting it to the Sp 2/0-Ag 14 Culture,
  Espoo: TKK/Automaatiotekniikan lab, 1992. (Helsinki University of Technology, Automation Technology Laboratory. Series A, Research reports, 7).

Journal article

• Arto Visala, Hannu Pitkänen and Aarne Halme,
  "Modeling of Chromatographic Separation Process with Wiener-MLP representation,"
  Journal of Process Control, Vol. 11, no. 5, pp. 443-458, 2001.
• X Zhang, A Visala, A Halme and P Linko,
  "Functional state modelling and fuzzy control of fed-batch aerobic baker's yeast process,"
  Journal of Biotechnology, Vol. 37, no. 2, 94-103, 1994.
• X Zhang, A Visala, A Halme and P Linko,
  "Functional state modelling approach for bioprocesses,"
  Journal of Process Control, Vol. 4, no. 3, 127-134, 1994.

Book article

• A Halme, A Visala and X Zhang,
  "Process modelling using the functional state approach,"
  in Multiple model approaches to modelling and control, T. Johansen R. Murray-Smith, Lontoo: Taylor & Francis, 1997, s. 121-143.
• A Visala, H Pitkänen and A Halme,
  "Wiener-type SOM- and MLP-classifiers for recognition of the dynamic modes,"
  in Lecture notes in computer science 1327, M. Hasler W. Gerstner and J.-P. Nicoud, Berlin: Springer, 1997, 1071-1076.

Conference article

• Janne Paanajärvi and Arto Visala,
  "Innovation form State-space Neural Network Model for Bioprocesses - Practical Considerations,"
  in The 8th Internationl Conference on Computer Applications in Biotechnology (CAB 8), 2001.
• A Visala,
  "Wiener -NN model and its use in modelling of biotechnical processes,"
  in Proc. Japan Finland 80th aniversary symposium on bioengineering & robotics, Tokyo, Japan, 27.4.1999, The Institute of Physical and Chemical Research, RIKEN, 1999.
• A Visala, J Paanajärvi, H Pitkänen and A Halme,
  "Wiener-tyyppisten NN-mallien käyttö prosessien valvonnassa,"
  in Automaatio 1999, Rauno Heinonen, Helsinki: Suomen Automaatioseura, 1999, s. 202-207.
• A Visala, H Pitkänen and J Paanajärvi,
  "Wiener-NN models and robust identification,"
  in Proc. 1999 International Joint Conference on Neural Networks, Washington DC, USA, 10.-16.7.1999, 1999.
• A Visala,
  "Identification of Wiener-MLP with feedback NOE-model with extended Kalman filter,"
  in Proc. IEEE World Congress on Computational Intelligence IJCNN, Anchorage, Alaska, 1998, IEEE, 1998, s. 1-6.
• A Visala,
  "Wiener-tyyppisten NN-mallien ominaisuudet ja käyttö,"
  in Proc. Oppivien ja älykkäiden järjestelmien sovellutukset, Espoo, 1998, Ossi Taipale, Lappeenranta: Teknologian kehittämiskeskus, 1998, 10 s.
• A Visala and A Halme,
  "Dynamic Wiener-MLP with feedback NOE-model for Tricoderma fermentation,"
  in Proc. The 7th International Conference on Computer Applications in Biotechnology, Osaka, Japan, 1998, Osaka: 1998, s. 37-42.
• A Visala and J Jumppanen,
  "Wiener-type dynamical NN-models,"
  in Oppivien ja älykkäiden järjestelmien sovellutukset, Espoo: Teknologian kehittämiskeskus, 1997, s. 1-6.
• A Visala, H Pitkänen and A Halme,
  "Describing the dynamics of a chromatographic separation process with Wiener-type NN-model,"
  in Proc. 10th international conference in aqueous two-phase systems, Reading, 1997, Reading: 1997, 4 s.
• A Visala and A Halme,
  "Quality assurance in bioprocesses by model based fault diagnosis and state estimation,"
  in Preprints of the 13th IFAC World Congress, IFAC, 1996, s. 425-430.
• A Visala and A Halme,
  "Recognition of functional states on the basis of Laguerre presentation,"
  in Proc. IMACS Symposium on Control, Optimization and Supervision, Lille, 1996, Ranska: IMACS, 1996, s. 263-268.
• A Halme, A Visala, J Selkäinaho and J Manninen,
  "Fault Detection by using Process Functional States and Temporal Logic Nets,"
  in Proc. Fault Detection, Supervision and Safety for Technical Processes, Espoo, 1994, London: Pergamon, 1994, 183-188.
• A Halme, A Visala and J Kankare,
  "Functional State Concept in Modelling of Biotechnical Processes,"
  in Proc. 2nd IFAC Symposium on Modeling and Control of Biotechnical Processes, Keystone, Colorado, 1992, Oxford: Pergamon press, 1992, 153-158.
• X Zhang and A Visala,
  "Functional States, Local Models and Estimation of Biomass in An aerobic Yeast Culture,"
  in Proc. Bioprocess Engineering Meeting, Stockholm, 1992, Lund: Biotechnology Research Foundation, 1992, 140-146.
• X Zhang, A Visala and A Halme,
  "A Kinetic Model of Mammalian Cell Cultures,"
  in Proc. 2nd IFAC Symposium on Modeling and Control of Biotechnical Processes, Colorado, 1992, Oxford: Pergamon press, 1992, 367-370.