OVERVIEW

This course introduces the key concepts related to hydrodynamic instability and turbulence. The study of turbulence will be based on the statistiacal properties of turbulent flow fields and on the Reynolds-averaged Navier Stokes equations. We will introduce computational fluid dynamics using the software Ansys Fluent, through classes and tutorials. Finally, we will introduce modern techniques in machine learning for simulating and navigating turbulent environments. 

AIMS AND CONTENT

LEARNING OUTCOMES

The aim of the course is to allow students to acquire a critical view of numerical strategies for the modelling of turbulence (both developed and in transition), both of RANS type and of LES type. The expected result is a consequent mature use of such numerical strategies, based on the awareness that simulations are not the real thing but a sophisticated modeling of it and, as such, susceptible to errors, sometimes relevant.

AIMS AND LEARNING OUTCOMES

​The course's main objectives are: 

• give students a thorough knowledge of turbulence modeling and CFD
• Analyze the properties of turbulence both with and without walls, with asymptotics and data analysis.
• Discuss background of RANS (Reynolds average Navier Stokes) models as well as DNS (direct numerical simulations) and LES (large eddy simulations).
• Analyze accuracy and reliability of CFD turbulent simulations, as well as discretization techniques, solution strategies, and best standard practices when conducting CFD simulations
• We also conduct hands-on sessions to help students develop essential skills to perform CFD simulations from scratch with Ansys Fluent.
• Acquire introductory notions on modern techniques in supervised learning and reinforcement learning to simulate and navigate turbulence.
• Manipulate simple Python codes, proposed by the teachers in the form of tutorials.  

PREREQUISITES

• Basic knowledge of Fluid Mechanics.
• Basic knowledge of numerical methods.

TEACHING METHODS

• The course is based on 60 hours of frontal lectures and hands on sessions 
• The course is graded in the base of continuous homework and a final project. There are no exams.
• Instructors can be contacted by the students for questions and clarifications during office hours.

Students who have valid certification of physical or learning disabilities on file with the University and who wish to discuss possible accommodations or other circumstances regarding lectures, coursework and exams, should speak both with the instructor and with Professor Federico Scarpa (federico.scarpa@unige.it), the Polytechnic School's disability liaison.

SYLLABUS/CONTENT

The program is divided between lectures and guided tutorials. We will also conduct numerical simulations or analyze data using modern software and applications.

1. Introduction to turbulence 
2. Statistical properties of turbulent flows
3. Length scales in turbulent flows. From Kolmogorov scales to Taylor microscales to integral scales. Energy cascade. Law of the wall. Near wall treatment.
4. Governing equations. Reynolds averaging. The Boussinesq hypothesis. Reynolds-averaged Navier-Stokes equations (RANS).
5. homogeneous isotropic turbulence
6. Wall bounded turbulence
7. Closure problem. Algebraic models. One equation models. Two equation models. Reynolds stress models (RSM). Unsteady RANS simulations (URANS). Wall modeling and wall resolving simulations.
8. Post-processing and analyzing turbulent simulations - Part 1. Statistical post-processing. Revisiting the concept of statistical description of turbulence. Descriptive statistics. Joint statistics. One-point correlation. Two-point correlations. Time series. Turbulent kinetic energy spectrum. Power spectrum.
9. Post-processing and analyzing turbulent simulations - Part 2. Quantitative and qualitative post-processing. Dealing with steady and unsteady simulations. Presenting results.
10. Beyond the Boussinesq hypothesis, compressibility effects, and multiphase flows. Effect of roughness on the law of the wall.
11. Scale-resolving simulations (SRS). DES, LES, DNS. Wall modeling and wall resolving simulations in SRS.
12. Best practices in CFD and turbulence modeling. Numerical considerations. Validation and verification. Mesh dependency studies. Accuracy and reliability of turbulent simulations.

RECOMMENDED READING/BIBLIOGRAPHY

Recommended literature

• D. Wilcox. Turbulence Modeling for CFD. DCW Industries Inc., 2010.
• S. Pope. Turbulent Flows. Cambridge University Press, 2000.
• P. Bernard. Turbulent Fluid Flow. Wiley, 2019.
• M. T. Landahl and E. Mollo-Christensen. Turbulence and Random Processes in Fluid Mechanics. Cambridge University Press, 1992.
• H. Tennekes and J. L. Lumley. A First Course in Turbulence. MIT Press, 1972.
• P.J. Schmid e D.S. Henningson, Stability and Transition in Shear Flows, Springer, 2001.
• U. Frisch, Turbulence, Cambridge Univ. Press. 1995.

TEACHERS AND EXAM BOARD

Exam Board

AGNESE SEMINARA (President)

JOEL ENRIQUE GUERRERO RIVAS

ANDREA MAZZINO (President Substitute)

LESSONS

LESSONS START

https://corsi.unige.it/en/corsi/9270/studenti-orario

EXAMS

EXAM DESCRIPTION

The final evaluation consists of a CFD project where the student must put into practice all the knowledge acquired. The case to be developed must be agreed upon between the examiner and the student.

The data and time of the final presentation must be agreed upon between the examiner and the student.

Students with specific learning disability (SLD), disability or other regularly certified special educational needs are advised to contact the instructor at the beginning of the course to agree on teaching and examination methods that, in compliance with the course objectives, take into account the individual learning requirements.

ASSESSMENT METHODS

• The course will be graded based on continuous assignments (no more than three) and a final project.
• A written report and a short presentation are expected at the end of the final project.
• Based on the examiner's feedback, the student will have a one-time opportunity to improve his/her final report.

FURTHER INFORMATION

Course website
http://www3.dicca.unige.it/guerrero/teaching_turbulence.html

For the hands-on sessions, students should bring her/his computer with all the software installed. All the software to be used is free and can be downloaded at the links provided below.

Required software:

Ansys Fluent student version (version 2021R1 and up) – CFD solver (only Windows)
https://www.ansys.com/academic/free-student-products

Optional software:

Anaconda Python (Python distribution 3.7) – Data analysis (and more).
https://www.anaconda.com/distribution/

Paraview (version 5.6 and up; however, I recommend version 5.6) – Scientific visualization
https://www.paraview.org/