Research scope
The research activities of the Process Systems Engineering (PSE) group address a wide variety of applications, both in the analysis and development of specific chemical processes and products, and in the development of new methodologies for systematic decision-making in process and product engineering:
The main areas of current research are the following:
Research activities involve the development of the optimization and supervision structures for 1st generation biodiesel processes, from used vegetal oils. The main goals are the improvement of the current transesterification dynamic models and the implementation of advanced control strategies in an industrial plant of a project partner. An additional undergoing effort regards the development of the concept of biorefineries, based both on lignocellulosic and oleaginous raw-materials. Here, previous efforts in the advancement of efficient purification technologies for large scale products (such as ethanol) will be continued, by considering alternative technical solutions (such as extractive fermentation coupled with liquid-liquid separation and the selection of new entrainers for extractive distillation). This study is coupled to the structure development for integrated processing steps that enable a more complete use of the initial raw-materials, with higher energy efficiency.
Another research line is the development of methodologies for chemical product design, and in particular the design of formulated and/or structured products and corresponding manufacturing processes. Various topics are being explored:
i. Conceptual developments, including overall methodologies, specific tools, pedagogic case studies, and the particular problem of microstructured products (how to systematically generate plausible alternatives).
ii. Systematic methods to select solvents and other auxiliary materials used in the production of structured products (e.g. solvents in micro-encapsulation processes, plasticizers in polymers production).
iii. Systematic study of attainable microstructures using spray drying as a base technology.
iv. Data gathering and analysis of industrial, academic and employment data relative to the practice of product engineering, and the elaboration of a set of recommendations and good practices for the teaching of Product Engineering.
v. Industrial control of bivalve invasive species. The objective is the development of improved methods (more efficient and more environmentally friendly) for the control of macro-biofouling in hydro-dependent industries.
vi. Development of methods for the eco-toxicological evaluation of structured products. Currently existing method should be extended to consider not only the presence of active ingredients, but also their micro (and nano) structure.
Several members of the group are actively developing process intensification methodologies applied to liquid-liquid systems in structured equipment. Here an experimental research facility will be used to evaluate the performance of micro- and meso-structured reactors, when processing liquid-liquid reactions. In parallel to the experimental program, mathematical models are being developed for experimental validation.
The development of detailed mathematical models for various chemical processes will be considered, as supporting tools for diagnosis, supervision and decision making. This task will be addressed considering a combination of data-driven and mechanistic approaches, and will address various specific applications such as aromatics hydrogenation and nitration, catalytic cracking (FCC) units, and the batch production of urea-formaldehyde resins.
A current research area is the design of nanoparticles of various oxide materials via gas-to-solid routes. Here, the main research purpose is the study of nanoparticles production employing detonation-based strategies, aiming to elucidating the mechanisms and phenomena involved by crossing simulation with experimental studies. Conceptual frameworks commonly employed in the PSE are will be used to structure knowledge, optimize the process/product and improve its controllability. Further research activities will also embrace alternative polymerization processes for biodegradable materials, for the production of poly-lactic acid.
In the area of computational statistics algorithms are being developed employing optimization based approaches. Among this group of problems is the optimal design of sampling plans and of plans of experiments for biomedical and social applications, both yielding classical bilevel programming and minimax problems. Optimization based approaches are also being developed for the solution of large-scale dynamic optimization problems, using an efficient simultaneous approach able to address real-time applications and the optimization of systems described by PDE models, in realistic applications. Another current activity that will also be addressed is the integration of advanced monitoring approaches with also advanced process control strategies (at the same level, in the hierarchy of process operations), such as model predictive control (MPC). By properly integrating both techniques, particular advantages of each one can be combined, leading to a global reduction of process induced variability and therefore to safer and more efficient processes.
Advanced control strategies for distributed parameter processes will be investigated. Nonlinear model predictive control of fedbatch beioreactors will be carried out in collaboration with the Service d'Automatique of the Polytechnique Faculty of the University of Mons. In the context of the monitoring, control and optimization of process operations, improved frameworks that address important features of typical systems will be developed. A previous research path in our research group will be followed, where methodologies were already developed for integrating the multivariate and multiscale nature of data, as well as its noisy/stochastic and dynamic structures. Here the explicit incorporation of systems network topology along with information about the causal directionality between connected elements will be considered, in data-driven analysis methods.