@article {232, title = {Global stabilisation of continuous bioreactors: Tools for analysis and design of feeding laws}, journal = {Automatica}, volume = {89}, year = {2018}, pages = {340 - 348}, abstract = {

This work revisits the dynamic behaviour of stirred continuous reactors in which a single bioreaction with unknown kinetics occurs. Conditions on the feeding strategy to avoid washing out the biomass and falling in batch operation are obtained. These conditions derive in a closed positively invariant region including the desired operating point. It is stated that no closed orbits may exist in this region and, furthermore, that no fixed point exists but on one of its borders. Therefore, global stability is achieved by finding a feeding law that fulfils the aforementioned invariant conditions and gives a single equilibrium for a first-order dynamics. These results are useful to determine the stability properties of different control laws and, more importantly, to design new ones. The main advantages of the proposed approach are its simplicity and that, differing from previous results, input saturation does not affect stability results. The potentiality of the developed tools is illustrated by means of classical and novel feeding laws.

}, keywords = {Bioprocess control, Invariant control, Nonlinear control}, issn = {0005-1098}, doi = {https://doi.org/10.1016/j.automatica.2017.12.041}, author = {De Battista, Hern{\'a}n and Jamilis, Mart{\'\i}n and Garelli, Fabricio and Pic{\'o}, Jes{\'u}s} } @article {48, title = {Second-order sliding mode observer for multiple kinetic rates estimation in bioprocesses}, journal = {Control Engineering Practice}, volume = {21}, year = {2013}, pages = {1259 - 1265}, abstract = {

Abstract Specific kinetic rates are key variables regarding metabolic activity in bioprocesses. They are non-linear functions of concentrations and operating conditions and therefore of difficult access for process control. In this paper, a multiple kinetic rates observer based on second-order sliding mode ideas is proposed. The main difference with other proposals is that smooth estimates are achieved in finite-time without adding additional dynamics. The resulting estimator is robust against uncertainty in the model of the estimated variables. Experimental results from continuous fermentation of S. cerevisiae are presented, where microbial specific growth rate and net ethanol production rate are estimated.

}, keywords = {Bioprocess control}, issn = {0967-0661}, doi = {http://dx.doi.org/10.1016/j.conengprac.2013.03.003}, url = {http://www.sciencedirect.com/science/article/pii/S0967066113000488}, author = {Nu{\~n}ez, Sebasti{\'a}n and De Battista, Hern{\'a}n and Garelli, Fabricio and Vignoni, Alejandro and Pic{\'o}, Jes{\'u}s} } @article {51, title = {Stability preserving maps for finite-time convergence: Super-twisting sliding-mode algorithm}, journal = {Automatica}, volume = {49}, year = {2013}, pages = {534 - 539}, abstract = {

The super-twisting algorithm (STA) has become the prototype of second-order sliding mode algorithm. It achieves finite time convergence by means of a continuous action, without using information about derivatives of the sliding constraint. Thus, chattering associated to traditional sliding-mode observers and controllers is reduced. The stability and finite-time convergence analysis have been jointly addressed from different points of view, most of them based on the use of scaling symmetries (homogeneity), or non-smooth Lyapunov functions. Departing from these approaches, in this contribution we decouple the stability analysis problem from that of finite-time convergence. A nonlinear change of coordinates and a time-scaling are used. In the new coordinates and time{\^a}{\texteuro}{\textquotedblleft}space, the transformed system is stabilized using any appropriate standard design method. Conditions under which the combination of the nonlinear coordinates transformation and the time-scaling is a stability preserving map are given. Provided convergence in the transformed space is faster than O(1/T) -where T is the transformed time- convergence of the original system takes place in finite-time. The method is illustrated by designing a generalized super-twisting observer able to cope with a broad class of perturbations.

}, keywords = {Bioprocess control}, issn = {0967-0661}, doi = {http://dx.doi.org/10.1016/j.conengprac.2013.03.003}, url = {http://www.sciencedirect.com/science/article/pii/S0967066113000488}, author = {Pic{\'o}, Jes{\'u}s and Pic{\'o}-Marco, Enric and Vignoni, Alejandro and De Battista, Hern{\'a}n} }