 | | |
| MECA0010-1 | Stochastic modelling in mechanics
|
|
| Duration : | 30h Th, 30h Pr |
|
| Number of credits : |
|
|
| Lecturer : | Maarten Arnst |
|
Language(s) of instruction :
|
| English language |
|
Course contents :
|
| The modeling and simulation of complex natural and engineered systems represent critical tools for addressing numerous science and engineering challenges. However, models are, by definition, only approximations of their target physical scenarios, and thus are prone to modeling errors; also, parametric uncertainties exist as a reflection of various limitations in experimental methods. The quantification and management of uncertainties thus constitute crucial requirements for models and simulations to find useful applications in supporting scientific discovery and engineering.
This course will introduce the fundamental concepts and methods underlying the quantification and management of uncertainties in predictive simulations.
- Fundamental concepts, including the distinction between parametric uncertainties and modeling errors, their relevance to scientific computation, and their treatment by probabilistic and other approaches.
- Representation of uncertainties, including the probabilistic characterization of uncertainties in a manner that is consistent with mechanics and physics.
- Propagation of uncertainties, including Monte Carlo sampling approaches.
- Management of uncertainties, including model validation and sensitivity analysis, with an emphasis on new engineering questions and approaches enabled by a detailed treatment of uncertainties.
Detailed examples of uncertainty analyses of relevant problems from aerospace and mechanical engineering, automotive engineering, climate change and next-generation renewable energy systems will be presented throughout the course. |
|
Learning outcomes of the course :
|
| At the end of the course, students will be able to identify the errors and uncertainties that may affect the predictions of models and simulations, and to put in place techniques for their quantification and management. Students will also learn how to read and understand relevant papers from the current and historical scientific literature. |
|
Prerequisites and co-requisites/ Recommended optional programme components :
|
| This course assumes that students have a background in elementary calculus (integrals and derivatives), linear algebra (matrices and vectors), probability and statistics (probability density functions, mean, variance) and mechanics and physics. |
|
Planned learning activities and teaching methods :
|
| The course takes the form of a series of lectures. Each lecture will be supported by slides prepared by the instructor, and revolve around a book chapter or paper from the scientific literature, which the students will have to read beforehand. In addition, students will work in small groups on a term project, which can take the form of a literature review or an implementation of an uncertainty analysis in Matlab. |
|
Mode of delivery (face-to-face ; distance-learning) :
|
| Face-to-face. |
|
Recommended or required readings :
|
| Each lecture will be supported by slides prepared by the instructor, and revolve around a book chapter or paper from the scientific literature, which will be made available through the course website. |
|
Assessment methods and criteria :
|
| Students will be required to prepare a classroom presentation of their term project, and document their work in a concise report. The final grade will be an equally weighted average of the grades obtained for the classroom presentation and the report, which will take into account their content, clarity, and neatness. |
|
Organizational remarks :
|
| The course will be offered in the Spring semester. |
|
Contacts :
|
| Dr. Maarten Arnst
Assistant professor
Bureau: B52 - 0/419
Email: Maarten.Arnst@ulg.ac.be |
|
