Sebastián Nuñez

Sufficient conditions for state observability in multi-substrate bioprocesses with additive growth dynamics

The distinguishability of the non-measured states is relevant for the design and implementation of state observers. The aim of this study is to examine the observability of a fed-batch multi-substrate process with additive growth rate. Based on the model equations, sufficient conditions to fulfill the observability rank condition were determined. Results for three sets of outputs were analyzed for Monod and Haldane models.

Specific Kinetic Rates Regulation in Multi-Substrate Fermentation Processes

Abstract The regulation of the biomass specific growth rate is an important goal in many biotechnological applications. To achieve this goal in fed-batch processes, several control strategies have been developed employing a closed loop version of the exponential feeding law, an estimation of the controlled variable and some error feedback term. Moreover, in some bioprocesses there is more than one feeding flow entering the bioreactor and supplying different nutrients or substrates. Hence, the problem of estimating multiple substrate consumption rates together with the specific growth rate of the microorganism becomes relevant. In this context, the dynamic behavior of fed-batch processes with multiple substrates and Haldane kinetics is further investigated. In particular, a nonlinear PI control law based on a partial state feedback with gain dependent on the output error is used. Then, with a recent developed algorithm for several kinetic rates estimation based on second-order sliding mode (SM) ideas, we extend the mentioned control strategy to a multi-substrate fed-batch bioprocess. The observer provides smooth estimates that converge in finite time to the time-varying parameters and allows independent design of the observer and controller dynamics. The features of the proposed estimation and control strategies are assessed by simulation in different scenarios.

Comprehensive analysis of a metabolic model for lipid production in Rhodosporidium toruloides

The yeast Rhodosporidium toruloides has been extensively studied for its application in biolipid production. The knowledge of its metabolism capabilities and the application of constraint-based flux analysis methodology provide useful information for process prediction and optimization. The accuracy of the resulting predictions is highly dependent on metabolic models. A metabolic reconstruction for R. toruloides metabolism has been recently published. On the basis of this model, we developed a curated version that unblocks the central nitrogen metabolism and, in addition, completes charge and mass balances in some reactions neglected in the former model. Then, a comprehensive analysis of network capability was performed with the curated model and compared with the published metabolic reconstruction. The flux distribution obtained by lipid optimization with flux balance analysis was able to replicate the internal biochemical changes that lead to lipogenesis in oleaginous microorganisms. These results motivate the development of a genome-scale model for complete elucidation of R. toruloides metabolism.