Ph. Bogaerts

Simple metabolic modelling of vero cell growth on glucose in fixed-bed bioreactors

Abstract A novel metabolic model is formulated to simulate Vero cell growth on glucose in fixed-bed bioreactors by simplification of glucose fluxes in the central carbon metabolism. Model formulation is based on three designated metabolism states: respiratory metabolism, overflow metabolism and critical metabolism. The model is validated by experimental culture results and successfully predicts the dynamics of cell growth, glucose consumption and lactate production. As well as quantitatively describing glucose overflow metabolism in mammalian cell culture, the proposed model will be helpful for directly monitoring and controlling metabolism state of cells in order to achieve process optimization.

Maximum likelihood parameter estimation of a hybrid neural-classical structure for the simulation of bioprocesses

This paper proposes a hybrid structure for the modeling of a bioprocess: classical (in the form of a priori knowledge describing the mass balances) and neural (a radial basis function network describing the nonlinear reactions kinetics within these mass balances). The aim is to build a continuoussimulator capable to reconstruct from initial conditions the trajectory of state variables (i.e. the main component concentrations) by considering also an aspect which usually is not taken into account in bioprocess modeling: the existence of important measurement errors. A clustering strategy is used for placing the Gaussian centers and a maximum likelihood cost function is defined for the estimation of the network weights and initial conditions for the simulator. The structure is tested on batch animal cell cultures for which rare and asynchronous measurements are available: glucose, glutamine, lactate and biomass concentrations.

Nonlinear and Linearized Full Horizon State Observers - Application to Bioprocesses

Abstract The state estimation of full horizon observers is based on the identification of the most likely initial state of the process, this latter being identified at each time where new measurements are available. On the one hand, a nonlinear version of this observer is given together with its properties. It has the advantage to consist of a real nonlinear approach (without any approximation of the nonlinear model of the process). On the other hand, a linearized version is proposed, based on a recursive most likelihood estimation of the initial state. This latter estimator does not need to solve a nonlinear optimisation problem at each new sample time anymore. These approaches are illustrated in the case of the biomass concentration estimation within CHO animal cell cultures, for which only rare and asynchronous measurement samples of the glutamine, glucose and lactate concentrations are available.

Stochastic full horizon – asymptotic hybrid observer applied to a simulated cell culture

Abstract In an attempt to combine the advantages of an exponential state observer (i.e. fast convergence with an accurate model) and an asymptotic one (i.e. convergence without any knowledge about the kinetic model), a stochastic hybrid observer, which compares the observations made by the two observers has been developed (Hulhoven et al ., 2003). This comparison is made in order to perform a test on the process model quality and to provide a state estimation that evolves, accordingly, between the state estimation provided by the exponential observer and the one from the asymptotic one. In this contribution, the stochastic hybrid observer is established by using a full horizon observer as the exponential observer. The performances of this hybrid observer, are then tested on a simulated fed-batch cell culture.

Biomass Reconstruction in a Wastewater Treatment Biofilter

Abstract this paper is concerned with a pilot-scale fixed-bed biofilter used for nitrogen removal from municipal wastewater. Process dynamics is described by a set of mass balance partial differential equations, which allow the evolution of the several component concentrations along the biofilter axis to be reproduced. Based on sets of experimental data collected over a several-month period, unknown model parameters are estimated by minimizing an output error criterion. The resulting distributed parameter model and a few pointwise measurements of nitrate, nitrite, and ethanol concentrations can be used to design observers, which allow the unmeasured biomass concentrations to be reconstructed on-line. First, it is demonstrated that asymptotic observers are unsuitable for the model structure. Then, a receding-horizon observer is designed and tested, which shows very satisfactory performance.

Modellidentifikation zur Zustandsschätzung – Anwendung auf einen Bioprozess (Model Identification for State Estimation Application to a Bioprocess)

Anhand eines Prozessmodells und einiger verfügbarer Messungen sind Zustandsbeobachter in der Lage, nicht gemessene Zustände online zu rekonstruieren. Wenn das zugrundeliegende Prozessmodell aufgestellt ist, werden die unbekannten Parameter gewöhnlich mit Hilfe der Methode der kleinsten Fehlerquadrate oder der Maximum-Likelihood-Methode identifiziert, die von Offline-Messungen des gesamten Zustandsvektors Gebrauch machen. Diese herkömmlichen Methoden bringen jedoch nicht zum Ausdruck, dass das Modell des nicht gemessenen Teils des Zustandsvektors mit dem gemessenen in einer sensitiven Abhängigkeit stehen soll. Um eine höhere Modellsensitivität zu erzielen, wird in dieser Studie eine neue Kostenfunktion vorgeschlagen, die das klassische Maximum-Likelihood-Kriterium mit einem skalaren Maß der Modellsensitivität kombiniert.

Identification for Fault Detection in an Industrial Condenser

Abstract The condenser of a batch distillation column, equipped with temperature and flow sensors, is considered. The aim of the paper is to build a system which is able to detect flow sensor failures occurring in a batch. Three methods are considered For all of them, the first step consists of the parameter identification of a grey box model. For the first two tests, the classical least squares approach is used The first fault detection test is based on the value of the sum of the squares of the prediction errors obtained with the current batch measurements. The second one is a classical hypothesis test relying on the log-likelihood ratio. Both methods are shown to lack robustness with respect to the process variability from batch to batch. A third approach is then investigated, in which during the identification phase, the optimised cost function is chosen in order to reflect the sensitivity of the model to the flow sensor fruits. It yields significantly better results than the other two methods for the considered data.