Context In Structural Health Monitoring (SHM), we want to automatically detect defects/damages on a structure in order to perform interventions. Usually the focus of SHM is the detection of changes in the modal characteristics of structures (i.e. eigenmodes and eigenvalues of the system). However, in order for a damage to...
[Read More]
Literature Graph Navigation using Pre-trained Transformers
Literature Clustering for Workshop I co-organized some small workshops in my department (Civil and Environmental Engineering) called “DSML@DBAUG” where researchers (Ph.D. and Post Docs) from all chairs of the civil and geomatic engineering at ETH shortly presented their work. Since this was happening during the pandemic, everything happened through zoom....
[Read More]
Graph Neural Networks and Wind Farms
Deep Neural Networks with Relational Inductive Biases for Wake Modeling
Wake effects
As discussed in my other post, wake effects are an important consideration when designing wind farms.
[Read More]
Single-channel localization with Deep CNNs
or what multiplexers don't want you to know
Motivation/context I was interested in figuring out an ultra-cheap way to transform my IKEA table to a touch screen or some sort of user interface. Figuring out the position of the table where a tap happens is the first step to doing that. This is the problem of acoustic source...
[Read More]
Conditional Variational Autoencoder for Wind Turbine Blade Fatigue Damage Estimation
The project
[Read More]