EspañolEnglish
Research Research projects Development of advanced tools for design methodologies and multi-objective optimization in control engineering. Application to multivariable systems.

Development of advanced tools for design methodologies and multi-objective optimization in control engineering. Application to multivariable systems.

Proyecto Retos I+D+I (DPI2015- 71443-R). Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016.
    |

SUMMARY:

The project aims, firstly, to develop advanced tools to apply the methodology of multi-objective optimization in the design of control systems for complex multivariable processes and, secondly, to advance in the development of multivariable controllers, making emphasis on model-based predictive controllers (widely used in process industry). Additionally, all results will be validated in relevant applications in various fields of engineering, especially energy and transport. In particular, developments will be tested on PEM fuel cells systems (both static and mobile applications) and issues related to the search for optimum operating points, energy management in vehicles, trajectory optimization, etc.
In its scientific aspect, the project will seek to provide innovative solutions in the three key components of a multi-objective optimization problem: at the objectives selection trying to bring them closer to the needs of the designer; at the optimization phase in improving algorithms to cope with a large number of variables and constraints; and at the phase of decision making, improving the incorporation of designer preferences and the multidimensional visualization.
Moreover, applying multiobjective tools to the design of multivariable controllers aims to achieve an intuitive and interpretable tuning, especially when control problems have many constraints. In particular, model based predictive control or MPC will be used as control methodology, which besides being a technology with great industrial acceptance, it keeps a close relationship with optimization. Similarly, given the importance of models and virtual sensors when you want to control complex multivariable processes, multiobjective methodology will also apply for achieving progress in the quality and reliability of dynamic models and the development of virtual sensors.


 

KEYWORDS: Multiobjective Optimization, Model Predictive Control, PEM fuel cell