Data-Driven Techniques: I have developed a framework for the design of optimal filters from data in linear and non-linear systems. The results have been obtained during my PhD and then in cooperation with professors Novara and Milanese from Politecnico di Torino. The proposed framework allows to estimate from data optimal or sub-optimal state filters for linear, LPV and non-linear systems. The methodology has been applied in automotive and power electronics applications. The present and future work is focused on the extension to direct controller design from data.
Demand response for Smart-Grids: I have developed a framework for the analysis of consumer behavior in demand response programs. The models are being employed in the design of novel strategies for demand side management. The results have been obtained with participation of PhD students at Universidad Javeriana. The present and future work is focused on the experimental evaluation of the strategies and the extension to distributed architectures, employing MPC approaches. Cooperation has been established with Prof. Kameshwar Poolla and Pravin Varaiya from UC Berkeley and G. Chicco from Politecnico di Torino.
Bioprocesses modelling and control: I have worked in cooperation with the Institute for the Study of Inborn Errors of Metabolism of U. Javeriana, in the development of models and control systems for the improvement of protein production in bioreactors, employing the Picha Pastoris microorganism. We have derived gray-box models of the microorganism’s metabolic processes and based on the models, we have constructed optimal growth strategies that have been successfully evaluated in the laboratory. The present work aims at developing a completely automated control system for the microorganism cultivation, controlling environmental variables and optimizing production.
