Sort by
Refine Your Search
-
to study and predict. In this four-year SNF-funded project, you will develop data-driven, multiscale simulation methods that combine computer simulations, machine learning, and surrogate models to explore
-
of measurement systems, signal processing and analysis and the assessment of measurement accuracy, robustness and long-term stability. The resulting data form the basis for model-based approaches to evaluating
-
system models to analyse whether spatially coherent urban and energy configurations can be operated efficiently under realistic physical, spatial, and infrastructural constraints. The work aims to identify
-
the use of hierarchical graph neural networks for modeling multi-scale urban energy systems. By combining advances in Physics-Informed Machine Learning (PIML) and Graph Neural Networks (GNNs) with real
-
activation and micromechanical modeling Progressive damage modeling of reinforced FRPs Mechanical characterization and fracture experiments Complete a PhD thesis at ETHZ Your profile Highly motivated and
-
of electron spectroscopy experiments Modelling of experimental data Your profile The position is immediately available for a candidate with a master's degree (or equivalent) preferably in physics, physical
-
approaches. The PhD will develop and apply optimization-based energy system models to analyse whether spatially coherent urban and energy configurations can be operated efficiently under realistic physical
-
and flow field interactions Tuning of the CFD models with experimental results Artificial Neural Network training and development Scientific publications in journals and at conferences Supervision
-
more cost-efficient. Together, UESL and IMOS are seeking a motivated and qualified PhD candidate to advance the use of hierarchical graph neural networks for modeling multi-scale urban energy systems. By
-
plumes from point sources using the MicroHH atmospheric model. Analysis of plume dynamics and NOx chemistry in the high-resolution simulations. Develop and refine data-driven methods for emission