Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Multiobjective Optimal Control: Algorithms, Approaches and Advice for the Application

: Gambier, A.


Institute of Electrical and Electronics Engineers -IEEE-:
International Automatic Control Conference, CACS 2020 : 4-7 November 2020, Hsinchu, Taiwan
Piscataway, NJ: IEEE, 2020
ISBN: 978-1-7281-7199-9
ISBN: 978-1-7281-7197-5
ISBN: 978-1-7281-7198-2
International Automatic Control Conference (CACS) <14, 2020, Hsinchu>
Fraunhofer IWES ()

The advanced control system design is turning more and more complex and, sophisticated optimization techniques become essential in order to satisfy such high requirements. In particular, multi-objective optimal (MOO) control is nowadays a way to reach a control system that conciliates at the same time several design objectives that sometimes could be contradictories. In the past ten years, numerous advances can be observed in the field of multi-objective optimization. Many optimization algorithms with high complexity are now available in order to solve different kind of optimization problems. The use of such approaches is not simple. Thus, the main contribution of this work is study of old and new MOO algorithms from the control point of view providing practical experience and design hints. A numerical study by using the pitch control system design of a wind turbine is included.