Online estimation of optimal operating conditions for simulated moving bed chromatographic processes.

A new approach for determining optimal operating conditions for simulated moving bed chromatographic processes is presented. The method is based on recursive online estimation and requires only rough initial estimates. It is based on a simple foot point model of the moving concentration fronts and an online measurement of the corresponding retention times in the different zones of the plant. A mathematical representation of the adsorption isotherms is not required. The method is validated experimentally for the separation of bicalutamide enantiomers.

Control and estimation of anaerobic digestion processes using hydrogen and volatile fatty acids measurements.

The anaerobic digestion (AD) technology is widely used in the treatment of waste and wastewater. To ensure the treatment efficiency and to increase the production of biogas, which can be reused as a renewable energy source, a good understanding of the process and tight control are needed. This paper presents an estimation and control scheme, which can be successfully used in the operation of the AD process. The process is simulated by the ADM1 model, the most complex and detailed model developed so far to characterize AD. The controller and the observer, which provides estimates of the unmeasurable variables needed in the computation of the control law, are designed based on a simplified model developed in a previous work. Since it has been shown that hydrogen concentration is an accurate and fast indicator of process stability, it was chosen as controlled variable. Aside from the hydrogen concentration, the only measurement employed by the proposed control structure is the volatile fatty acids concentration. Simulation results prove the effectiveness of the proposed control structure.

Parametric uncertainties and influence of the dead volume representation in modelling simulated moving bed separation processes.

In this study, a systematic numerical procedure for identifying the model parameters of simulated moving bed (SMB) separation processes is developed. The parameters are first estimated by minimizing a weighted least-squares criterion using experimental data from batch experiments, e.g. the time evolution of the concentration of elution peaks. Then, a cross-validation is achieved using data from experiments in SMB operation. At this stage, the importance of a careful modelling of the dead volumes within the SMB process is highlighted. In addition, confidence intervals on the estimated parameters and on the predicted concentration profiles are evaluated.

A detailed metabolic flux analysis of an underdetermined network of CHO cells.

In this article the metabolic flux analysis of growing CHO-320 cells is performed for a detailed metabolic network which involves 100 reactions and embraces all the significant pathways describing the metabolism of CHO cells. The purpose is to investigate the efficiency of the flux analysis when it is based on a relatively small set of extracellular measurements that can be easily achieved in most laboratories. In this case the flux analysis problem leads to a generally underdetermined mass balance system, as data are not sufficient to uniquely define the metabolic fluxes. Our main contribution is to show that, provided the system of mass balance equations is well-posed, although it is underdetermined, very narrow intervals may be found for most fluxes. The importance of checking the well-posedness of the problem is emphasized and the influence of the number of available measurements on the accuracy of the metabolic flux intervals is systematically investigated. In all cases the computed flux intervals are bounded and a single well defined value is obtained for the formation rates of the cellular macromolecules (proteins, DNA, RNA, lipids) that are not measured. The potential gain of a simple theoretical assumption regarding the metabolism of Threonine is also discussed and compared with an optimal solution calculated by maximizing the biomass formation rate. Alternative network structures obtained by inverting the direction of reversible reactions are also considered. Finally, the results of the metabolic flux analysis are exploited to estimate the total energy production resulting from the metabolism of growing CHO-320 cells.

Development and calibration of a nitrification PDE model based on experimental data issued from biofilter treating drinking water.

To remove ammonia for production of drinking water, nitrification can be performed in a bio-filter. At least 1 month is necessary to capture from the groundwater and then grow a sufficient amount of nitrifying bacteria to reach the desired removal efficiency. Improving start-up of bio-filters at low substrate concentration is therefore a major challenge. In this connection, it is important to develop appropriate models for designing, monitoring or analysing biofilm systems during start-up or following disinfection events. This study discusses the development and calibration of a nitrification PDE model which reflects the compromise between the complexity associated with the description of the full physical and biochemical mechanisms and the search for a simplified model with identifiable parameters. This model takes only the relevant phenomena (considering the full operating range) into account. The validity of the calibrated model has been evaluated through experiments under very different operational conditions, at the laboratory and under real industrial conditions, involving the full upstream chain of water treatment (iron oxidation and sand filter).