Student Projects

Student projects at IFD are usually closely related to current research projects and are supervised by our doctoral students, lecturers and professors. Some projects are conducted in collaboration with academic and industrial partners.

The Process

At IFD we try to assign the projects according to the students' preferences. To make sure that you get the project of your choice, it is best to contact us as soon as possible and preferably a few weeks before the semester starts. For a list of available projects, see the list below.

On the first Friday of each semester, IFD organizes an information event for students who are starting a project at IFD. The event is a great opportunity to meet other students, IFD faculty members, and supervisors. Moreover, detailed information about how student projects are conducted at IFD is given (Download student project guidelines (PDF, 3.3 MB)). Specifics about the event are provided by student supervisors.

Presentations of Bachelor, semester and CSE seminar thesis projects usually take place during the last week of the semester in the room ML H 51 ("Treibhaus"). Master thesis presentation dates are setup individually depending on the corresponding starting dates. Selected posters of projects are showcased on the H-floor of the ML building (poster templates for Download LaTeX (ZIP, 551 KB) and Download MS Word (DOCX, 353 KB) are available).

Available Projects

Subharmonic emissions from acoustically-coupled microbubble clouds

Group Supponen

In the medical context, coated microbubbles have the potential to measure the cardiac pressure noninvasively by means of subharmonic emissions. In this study, we aim to study the influence of acoustic coupling between neighboring microbubbles in the amplitude of the subharmonic emissions.

Keywords

Microbubble dynamics, numerical, subharmonic, pressure estimation.

Labels

Semester Project , Bachelor Thesis , ETH Zurich (ETHZ)

PLEASE LOG IN TO SEE DESCRIPTION

More information

Open this project...

Published since: 2025-12-19 , Earliest start: 2026-01-05

Applications limited to ETH Zurich

Organization Group Supponen

Hosts Collado Gonzalo

Topics Engineering and Technology , Physics

Numerical investigations of the nonlinear coupling between the shape modes and volumetric oscillations of microbubbles

Group Supponen

Microbubble dynamics are rich and complex. When they are excited with an ultrasound wave, these can undergo spherical and non-spherical oscillations. Experimental results suggest that non-spherical shape modes can couple with the spherical oscillation, leading to some unexpected behaviors. In this project we aim to numerically investigate this phenomenon.

Keywords

Microbubble, dynamics, partial differential equations, numerical

Labels

Semester Project , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)

PLEASE LOG IN TO SEE DESCRIPTION

More information

Open this project...

Published since: 2025-12-19 , Earliest start: 2026-01-05

Applications limited to ETH Zurich

Organization Group Supponen

Hosts Collado Gonzalo

Topics Mathematical Sciences , Engineering and Technology , Physics

Exploiting data-driven aerodynamic models to predict the dynamic response of an elastic wing

Institute of Fluid Dynamics

The Fourth Aeroelastic Prediction Workshop (AePW4) provides an open and impartial forum to assess and evaluate the current state of the art and state of the practice in computational aeroelastic modeling. This includes evaluating how well existing tools predict aeroelastic phenomena critical to aircraft analysis and design (flutter, buffet, structural deformation, supersonic aeroelastic effects, etc.).

Labels

Semester Project

Description

As a function of Mach number and angle of attack, shock waves may appear on either side of the wing and cause boundary-layer separation. The Baseline Supercritical Wing (BSCW) model considered here is elastically mounted to the wind tunnel wall using a special mechanism that allows it to translate vertically (plunge) and rotate (pitch). The interaction between the model kinematics and the transonic aerodynamics results in an aeroelastic system characterized by a strong nonlinearity that depends on the combined effects of Mach number and angle of attack. While the dynamic response is relatively easy to predict at Mach numbers below 0.7 and angles of attack below 2 degrees, under other conditions the aerodynamics are so strongly affected by shock–boundary-layer interaction that no conventional RANS CFD solver has yet provided a fully reliable prediction. This project focuses on the use of data-driven aerodynamic models and exploits high-quality CFD data that already exist. On the one hand, this type of approach is several orders of magnitude faster than a full CFD model. On the other hand, it may introduce errors due to data quality or the chosen data-generation method. However, owing to its simplicity, the effects of these errors can be readily propagated into the predictions, providing a framework to understand the relative importance of both the CFD errors and the reduced-model generation errors. More specifically, the project is composed of the following tasks: - Become familiar with the specific transonic flow conditions associated with the High-Angle Working Group’s interests, namely the BSCW wind-tunnel model. Review the existing literature and focus on the reasons behind the failure of CFD to correctly predict flutter speed. - Become familiar with the most relevant numerical approaches for aeroelastic flutter identification, including full models, linearized approaches, and methods based on aerodynamic reduced-order models and data-driven surrogate models. - Identify a promising data-driven approach and develop a methodology suitable for generating data-driven models in a two-parameter space (Mach number and angle of attack). Sufficient high-quality CFD data (in terms of dynamic responses to prescribed input signals) will be used to validate the approach. - Compute flutter speed for a large number of points in the parameter space, following either Mach-number sweeps or angle-of-attack sweeps. - Quantify the uncertainty associated both with the CFD data and with the data-driven model generation, ideally by defining a confidence interval. - Documentation and presentation of the project results.

Contact Details

More information

Open this project...

Published since: 2025-12-15

Applications limited to ETH Zurich

Organization Institute of Fluid Dynamics

Hosts Meyer-Massetti Daniel Werner

Topics Engineering and Technology

Flow simulation and uncertainty quantification of wing flutter

Institute of Fluid Dynamics

Labels

Semester Project

Description

AePW4 is divided into four Working Groups. The Large-Deflection Working Group (LDWG) deals with slender wings or structures undergoing large nonlinear structural deformations due to aerodynamic loading. In practice, this Working Group focuses on the dynamic stability of a wind-tunnel model called the Pazy wing, which is a rectangular (or trapezoidal, if built with a sweep angle) slender wing with an aspect ratio of 11. This model is designed to be built easily using readily accessible materials and tools. The existing literature shows that the flutter analysis of the straight Pazy wing can be performed using simple (low-fidelity) aerodynamic and structural models. The Pazy wing may undergo dynamic instabilities at wind speeds as low as 30 m/s. While the flutter mechanism of the straight version of the wing represents a classical flutter case, the behavior of the same wing becomes more difficult to analyze when it is built with a sweep angle. This project focuses on two main aspects: - The modeling of the swept Pazy wing, and - The analysis of the associated uncertainties. While the aerodynamic modeling effort may rely on low-fidelity aerodynamics (at a minimum, strip theory coupled with Theodorsen’s theory), the structural model may require more detailed attention. The uncertainty analysis entails both the characterization of aleatory uncertainty (associated with variability in the flow conditions and wind-tunnel model properties) and epistemic uncertainty (associated with the model itself). This may be carried out, for instance, using Polynomial Chaos Theory. More specifically, the project consists of the following tasks: - Become familiar with the flutter mechanism associated with the Pazy wing and review the existing literature. - Identify the most relevant modeling approaches for wings with comparable aspect ratio and flexibility. In particular, evaluate existing aerodynamic modeling approaches and characterize the uncertainties (due to modeling errors) associated with each of them. - Identify a suitable methodology and develop a model for both the straight and swept Pazy wing. A working low-fidelity model for the straight Pazy wing is already available. - Determine flutter speed for a suitable interval of angles of attack, typically from 0 to 10 degrees. - Quantify the uncertainty associated with the data (aleatory) and with the model error (epistemic), ideally by defining confidence intervals. - Documentation and presentation of the project results.

Contact Details

More information

Open this project...

Published since: 2025-12-15

Applications limited to ETH Zurich

Organization Institute of Fluid Dynamics

Hosts Meyer-Massetti Daniel Werner

Topics Engineering and Technology

Propagation-Based Phase-Contrast Imaging with Cold Neutrons: Towards Tomographic Imaging of Weakly Absorbing Metallic Objects

Group Supponen

Neutron imaging provides unique opportunity for studying technologically and biologically relevant materials as a complementary modality to other techniques. The increased interest in neutron imaging relies on their strong penetration into metallic materials, their sensitivity to light elements, and the possibility of manipulating contrast via isotope substitution. Among the imaging approaches, propagation based phase-contrast imaging recently showed a promising extension to conventional absorption transmission imaging. This method, well established with synchrotron X-ray radiation, relies on a (partially) coherent beam and carefully controlled experimental conditions to convert phase variations of the incoming wavefront into quantitative measurements. Note that this opportunity is for the Institut Laue-Langevin in Grenoble, France.

Keywords

phase-contrast imaging; synchrotron radiation; imaging techniques

Labels

Internship

Description

Goal

Contact Details

More information

Open this project...

Published since: 2025-12-12 , Earliest start: 2026-01-01 , Latest end: 2026-12-31

Organization Group Supponen

Hosts Brewer Cameron

Topics Engineering and Technology

Snow Saltation Measurements in the Field

Group Coletti

Saltating snow is an important phenomenon of mass transport over the snow cover, impacting various processes governing the snowpack evolution. Key to the accurate modelling of snow saltation is the mass flux of snow and shear distribution under drifting and blowing snow conditions. This project will consider field measurements of snow saltation outdoors in the Swiss Alps. The student will build and reproduce an existing snow saltation tracking system using available camera equipment and collocate measurements with a new outdoor holography setup to perform airborne particle characterization. This will acquire novel field data to study mechanisms of snow saltation coupled to sublimation in the near-ground surface layer.

Keywords

Snow Saltation, Drifting and Blowing Snow, Sublimation, Holography, Particle Tracking, Field Measurements

Labels

Semester Project , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)

Description

Goal

Contact Details

More information

Open this project...

Published since: 2025-11-25 , Earliest start: 2025-08-08 , Latest end: 2026-05-31

Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne

Organization Group Coletti

Hosts Muller Koen , Coletti Filippo

Topics Information, Computing and Communication Sciences , Engineering and Technology , Earth Sciences , Physics

3-Dimensional Convolutional Neural Network for semantic segmentation

Group Supponen

X-ray micro-computed tomography is an imaging technique enabling interior characterization of material micro-structures. The incident waves penetrate through the sample and are attenuated by the material density to be absorbed by a detector, represented by a quantized greyscale value. By rotating the sample and iteratively recording radiographic projections, internal boundaries between material phases can be retrieved. To recover quantifiable information regarding the sample, it is crucial to accurately perform semantic segmentation on the computed tomographs. Semantic segmentation refers to the categorization of each individual pixel within an image (voxel in 3D) into bins representing the various respective materials within the original data. Semantic segmentation poses relevance to a wide variety of computer vision and image analysis fields including autonomous driving, biomedical diagnosis, and agricultural analytics. Typical semantic segmentation often utilizes regional texture and color analysis or pixel distribution methods (Gaussian mixture and K-means) to determine pixel classification. These methods are highly reliable yet often are not time efficient on large datasets. Additionally, specific tuning is required for differing samples.

Keywords

Machine Learning; X-ray imaging; Material characterization

Labels

Semester Project

Description

Goal

Contact Details

More information

Open this project...

Published since: 2025-11-10 , Earliest start: 2026-01-12 , Latest end: 2026-05-29

Organization Group Supponen

Hosts Brewer Cameron

Topics Engineering and Technology

K-wave simulation of non-linear effects on HIFU transducer wave propagation

Group Supponen

High intensity focused ultrasound is a non-invasive biomedical treatment for tissue ablation. Through the propagation of high intensity pressure waves, tissue material can be destroyed by thermal effects or cavitation damage in a pulsed signal. The approach is utilized in treatment of tumours in many parts of the body including breast, kidney, liver, and prostate. The high intensity of the incident wave leads to a non-constant speed of sound within the medium generating non-linearity effects on the wave. Wave non-linearity inherently shifts wave energy to higher harmonics, influencing bubble stability and cavitation threshold. A past student project has used K-wave (a Matlab package) to simulate the pressure field of an ultrasound transducer, focusing on signal reflection and distortion resulting from impeding objects within the propagation field. This project builds upon the previously completed work by incorporating non-linearity in the simulation focusing on high intensity burst waves and enables efficient transducer design by simulating the change in the pressure signal as the number of transducer elements and configuration are manipulated.

Keywords

Nonlinear acoustics; Numerical simulation; Code optimization

Labels

Semester Project

Description

Goal

Contact Details

More information

Open this project...

Published since: 2025-11-10 , Earliest start: 2026-01-12 , Latest end: 2026-05-29

Organization Group Supponen

Hosts Brewer Cameron

Topics Engineering and Technology

Shape oscillations of bubbles on an oscillating surface

Group Supponen

When a microbubble resting on a surface is excited around its resonance frequency, the bubble exhibits intense shape deformations. This project aims at designing an experimental setup to capture the surface instabilities of accelerated bubbles.

Keywords

Fluid dynamics, bubbles, instability

Labels

Semester Project , Bachelor Thesis , ETH Zurich (ETHZ)

PLEASE LOG IN TO SEE DESCRIPTION

More information

Open this project...

Published since: 2025-11-06 , Earliest start: 2026-01-01 , Latest end: 2026-06-30

Applications limited to Department of Mechanical and Process Engineering

Organization Group Supponen

Hosts Presse Louan

Topics Engineering and Technology

Microbubble characterization through acoustic radiation force

Group Supponen

Acoustically driven microbubbles could unlock local drug delivery in a clinical setting. This therapy requires a precise characterization of the microbubble volumetric oscillations. This project will investigate an alternative method to characterize the volumetric oscillations indirectly through the acoustic radiation force. For this, the student is expected to conduct experiments and compare them with the predictions from theoretical models.

Keywords

Biomedical, drug delivery, ultrasound contrast agents, physics, experimental

Labels

Semester Project , Master Thesis

PLEASE LOG IN TO SEE DESCRIPTION

More information

Open this project...

Published since: 2025-10-28 , Earliest start: 2025-12-01

Applications limited to ETH Zurich

Organization Group Supponen

Hosts Collado Gonzalo

Topics Engineering and Technology , Physics

Experimental Visualization of Phase Shift in Focused Ultrasound

Group Supponen

Keywords

Ultrasound, Phase Shift, Ultrasound Transducers, Acoustics, Wave Physics

Labels

Semester Project , Internship , Master Thesis , ETH Zurich (ETHZ)

PLEASE LOG IN TO SEE DESCRIPTION

More information

Open this project...

Published since: 2025-09-29 , Earliest start: 2025-09-30 , Latest end: 2026-09-30

Organization Group Supponen

Hosts Fiorini Samuele

Topics Mathematical Sciences , Engineering and Technology , Physics

Digital twin of fetal aortic hemodynamics (Collaboration - Aarhus University, Denmark)

Group Coletti

Congenital heart defects (CHDs) affect a wide range of society around the globe with a prevalence of 1.0–1.2%. Hemodynamic forces such as the blood flow driven shear stresses are critical parameters that can influence the proper development of the endothelial and endocardial tissues. The hemodynamic development in the left heart is particularly important because the left heart provides the required pumping power for the blood circulation. In this extremely novel and pioneering project, we are employing MRI acquisitions of fetuses at different gestational ages for building digital twins of cardiovascular development. Such a model has the potential to revolutionize diagnostic procedures and provide invaluable tools for personalized medicine.

Keywords

Computational modelling; clinical imaging; hemodynamic modelling

Labels

Semester Project , Master Thesis

Description

Congenital heart defects (CHDs) affect a wide range of society around the globe with a prevalence of 1.0–1.2%. It is reported that only 2–4% of CHDs are observed in families with prior CHD history, showing that most of the cases are observed in infants with no prior history of disease in their families. This fact indicates that the mechanobiological mechanisms are also important in the formation of CHDs. Hemodynamic forces such as the blood flow driven shear stresses are critical parameters that can influence the proper development of the endothelial and endocardial tissues. The hemodynamic development in the left heart is particularly important because the left heart provides the required pumping power for the blood circulation. However, clinical in vivo flow measurements and hemodynamic assessments are quite challenging for the fetal hearts. In this extremely novel and pioneering project, thanks to the collaboration with the advanced research group at Aarhus University Hospital (Prof. Dr. Anne Skakkebæk and team), we are employing MRI acquisitions of fetuses at different gestational ages for building digital twins of cardiovascular development. Such a model has the potential to revolutionize diagnostic procedures and provide invaluable tools for personalized medicine. **Learning Outcomes**: This project offers opportunities to gain expertise in cutting-edge areas, such as: - Advanced computational fluid dynamics and cardiovascular flow analysis. - Mastery of clinical imaging segmentation techniques for CAE applications. Students will gain advanced knowledge in computational fluid dynamics (CFD), biomechanical device modeling, and the application of clinical imaging. This project emphasizes the use of industry-standard simulation software, such as COMSOL, alongside CAD modeling tools, to create precise and realistic simulations. **Preferred Prerequisites**: Candidates should have a strong foundation in fluid mechanics, CAD modeling, and experience with CFD or finite element analysis (FEA). Additional knowledge of biofluid dynamics, medical device design. R&D projects will be beneficial but not mandatory. **Impact and Contribution**: The integration of CFD, clinical imaging, and data-driven modeling in this project could significantly advance predictive diagnostics for fetal cardiovascular diseases.

Goal

Contact Details

More information

Open this project...

Published since: 2025-08-20 , Earliest start: 2025-09-01

Applications limited to ETH Zurich

Organization Group Coletti

Hosts Coletti Filippo , Li Yinghui

Topics Engineering and Technology

Surgical treatments of heart valve insufficiency: computational optimization through in-vitro and clinical data (Collaboration - Aarhus University, Denmark)

Group Coletti

Innovative approaches are needed for understanding and treating complex cardiovascular diseases. In this context, the amalgamation of in-silico methods, flow-loop experiments and 4D imaging emerges as a transformative paradigm in cardiovascular research.

Keywords

Computational modelling; MRI 4D-flow validation; fluid dynamic flow-loop; biofluid dynamics

Labels

Semester Project , Master Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2025-08-20 , Earliest start: 2025-09-01

Applications limited to ETH Zurich

Organization Group Coletti

Hosts Coletti Filippo

Topics Engineering and Technology

Fluid-structure interaction (FSI) modelling of tri-leaflet valves in cardiovascular applications (Collaboration - Aarhus University, Denmark)

Group Coletti

In-silico approaches such as computational fluid dynamics (CFD) and fluid-structure interaction (FSI) analyses, offer a comprehensive platform for predicting disease progression and designing medical devices, reducing the reliance on costly and invasive experiments. In this translational project, we aim at developing complex numerical models of tri-leaflet aortic valves, and at validating them with clinical measurements provided by collaborators at AUH.

Keywords

Computational modelling; cardiovascular engineering; 4D magnetic resonance imaging

Labels

Semester Project , Master Thesis

Description

Goal

Contact Details

More information

Open this project...

Published since: 2025-08-20 , Earliest start: 2025-09-01

Applications limited to ETH Zurich

Organization Group Coletti

Hosts Coletti Filippo

Topics Engineering and Technology