Student Assistant ​/ Master Thesis - Deep Learning for Urban Climate Emulation

Traditional CFD solvers are too slow for iterative design applications, requiring days to evaluate a single urban design variant. Fraunhofer IBP is pioneering the use of Deep Learning emulators to provide high-fidelity microclimate predictions at speeds that are much higher than physics-based numerical solvers. We are seeking a Master’s student with a background in Scientific Computing or AI to develop neural operators that utilize Signed Distance Functions for geometry-resolving urban physics.

• Assist with the generation of synthetic urban datasets using large-eddy simulations (LES) to serve as training and validation data.
• Implement Signed Distance Functions (SDF) and their gradients to represent complex building geometries as inputs to the neural network.
• Develop and train Deep Learning (DL) models to approximate 3D wind and temperature fields in urban districts.
• Evaluate the model’s generalization capabilities.
• Compare the accuracy and inference speed of the DL emulator against traditional CFD results.

• Enrollment in Scientific Computing, Computer Science, Physics, Geoinformatics, Meteorology, or a related computational field.
• Strong programming skills in Python and deep learning frameworks (PyTorch or Tensor Flow), or a strong willingness to learn.
• Interest in geometric deep learning and the physics of fluid dynamics, or a strong willingness to learn.
• Familiarity with 3D data representations (voxels, point clouds, or SDFs), or a strong willingness to learn.
• Strong mathematical foundation in Partial Differential Equations (PDEs) and Fourier analysis.

• A Master’s thesis opportunity (6-12 months, or longer) at the intersection of AI and urban physics.
• Access to high-performance workstations and GPU resources for training large-scale models.
• Mentorship from experts in both atmospheric modeling, meteorology and machine learning.
• Access to a collaborative platform linking Fraunhofer IBP in Germany and Concordia University in Canada.

The weekly working time is at least 5 hours. This position is limited, an extension possible. We value and promote the diversity of our employees' skills and therefore welcome all applications – regardless of age, gender, nationality, ethnic and social origin, religion, ideology, disability, sexual orientation and identity. Severely disabled persons are given preference in the event of equal suitability. Our tasks are diverse and adaptable – for applicants with disabilities, we work together to find solutions that best promote their abilities.

Remuneration according to the general works agreement for employing assistant staff.

With its focus on developing key technologies that are vital for the future and enabling the commercial utilization of this work by business and industry, Fraunhofer plays a central role in the innovation process. As a pioneer and catalyst for groundbreaking developments and scientific excellence, Fraunhofer helps shape society now and in the future.

If you have any questions, you may contact:

Dr. Afshin Afshari

Phone:

Fraunhofer Institute for Building Physics IBP

Requisition Number: 85181 Final date to receive applications: