101-0008-00L Identification Methods for Structural Systems
| Semester | Spring Semester 2018 |
| Lecturers | E. Chatzi, V. Ntertimanis |
| Periodicity | yearly recurring course |
| Course | Does not take place this semester. |
| Language of instruction | English |
| Abstract | This course will present methodologies for defining a structural system, and assessing its condition based on structural response data. This data is made available via measurements, which are nowadays available from low-cost and easily deployed sensor technologies. The course will explain how engineers may exploit technology for designing and maintaining a safe and resilient infrastructure. |
| Objective | This course aims at providing a graduate level introduction into the modeling and identification of structural systems. The goal is to establish relationships governing the system behavior and to identify the characteristics (mechanical, geometrical properties) of the system itself, based on noisy or incomplete measurements of the structural response. The course will include theory, as well as laboratory and actual-scale structural testing, thereby offering a well-rounded overview of the ways in which we may extract response data from structures. |
| Content | The topics to be covered are : - Fundamentals of vibrational analysis, signal processing and structural system representation - Modal Testing, Operational Modal Analysis - Parametric & Nonparametric Identification: Frequency Domain decomposition, Least Squares methods, ARMA models, Bayesian approaches. - Heuristic methods: Genetic Algorithms, Neural Networks. The differences between linear and nonlinear system identification will also be addressed. A comprehensive series of computer/lab exercises and in-class demonstrations will take place, providing a "hands-on" feel for the course topics. Grading: The final grade will be obtained, either - by 30% from the graded exercises and 70% from the written session examination, or - by the written session examination exclusively. The highest ranking of the above two options will be used, so that assignments are only used to strengthen the grade. |
| Lecture notes | The course script is composed by the lecture slides, which are available online and will be continuously updated throughout the duration of the course: Link |
| Literature | Suggested Reading: T. Söderström and P. Stoica: System Identification, Prentice Hall International: http://user.it.uu.se/~ts/sysidbook.pdf |