Data Science in Fluid Mechanics

Lecture:  
Date and time: Monday, 13:15 - 14:45
Location: Lecture room at the Institute of Fluid Mechanics
Begin: 08.04.2024
Lecturer: Dr.-Ing. A. Bauknecht
  The lecture is held in english language.
  For acceptance of the module "Data Science in Fluid Mechanics" you also have to participate in the exercise sessions.

Exam dates

 

Exercises  
Date and time: Monday, 15:00 - 16:30
Location: CIP pool at the Institute of Fluid Mechanics
Begin: 15.04.2024
Contact: Dr.-Ing. A. Bauknecht

Course objective

The course teaches students how to analyze and assess flow data from different sources, e.g. numerical (CFD) and experimental (PIV, hot-wire, etc.) approaches. At the end of the course, the students should be able to quantify how good their data are, to extract the main flow features and visualize them. In particular, the students should be able to assess the statistical and geometric error of their data for the first (mean) and second order (Reynolds stresses) quantities. They will also learn how to infer the error sources and their propagation. The students will be able to visualize and detect vortices and vorticity fields. Finally, the students will analyze the spectral content of their data using Fourier transform and dynamic mode decomposition, and extract coherent structures using proper orthogonal decomposition.

Course content

  • Fourier transform, Correlation function and spectra
  • Statistical principles, Statistical error, Geometric error, Estimator, expectation, variance and variability
  • Propagation of error
  • Vortex detection
  • Proper orthogonal decomposition, Dynamic mode decomposition