Abstract
Given the limited time an Academic Institution is given to educate engineering students and the ever expanding amount of knowledge, students are moreand more forced to choose between an early specialized in-depth engineering
track or a general but more superficial education. The result is that students
either graduate as specialists in one discipline or become generalists with a
more general knowledge in two or more disciplines.
Unless the students choose a specialization in system identification, they only
master Linear Time-Invariant system concepts and identification at best. This
general knowledge explains the basic linear time invariant system concepts and
introduces basic identification tools, but often falls short to tackle real world
systems that often possess a more rich behavior. Applying basic techniques
to real systems therefore often results in poor models, hampering the development and design of new products, while suitable methods exist but remain out of reach for many potential users.
The goal of the PhD is to soften the learning curve for more advanced methods to improve their use in practical conditions. We design a self-study environment to provide practitioners within industry and non-identification professionals with a more practical and custom tailored learning approach for
system identification, with a focus put on hands-on training. We believe applying new tools and concepts directly to real world toy systems can bridge the
gap between the newly gathered abstract knowledge and practically applicable methods ready for real industrial use. We believe that academic students
also benefit from such an approach as it sharpens their practical skills and
prepares them better for our model-driven engineering world.
| Date of Award | Mar 2019 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Francesco Ferranti (Jury), Frans Verbeyst (Jury), Tim De Troyer (Jury), Heidi Ottevaere (Jury) & Roger Vounckx (Jury) |