projects and literature

This commit is contained in:
Steffen Illium
2023-12-15 17:25:23 +01:00
committed by Steffen Illium
parent 86228fbaf7
commit b1a40905da
29 changed files with 67 additions and 4 deletions

View File

@ -10,3 +10,34 @@ header:
---
![Leaking pipe image](/assets/images/projects/pipe_leak.png){: .align-left style="padding:0.1em; width:5em"}
In cooperation with the Munich City Services ([Stadtwerke München (SWM)](https://www.swm.de/)), we researched the possibilities of leakage detection in real-world water networks.
Therefore, we installed contact microphones directly on the infrastructure in one of Munich's suburbs. {% cite muller2021acoustic%}
<center>
<figure class="half" style="max-width: 70%; text-align:center;">
<img src="\assets\images\projects\pipe_leak\1fe8265e-ff21-4e9c-8a2a-2ebcaec41728.jpeg" style="margin-bottom: 0em;">
<img src="\assets\images\projects\pipe_leak\8d2364f1-1b03-480d-9ed3-09d548f47383.jpeg" style="margin-bottom: 0em;">
</figure>
</center><br>
The feasibility study revealed the technical limitations of this undertaking. Nonetheless, we gain valuable insights. Through audio-to-mel spectrogram transformations and the power of machine learning, deep neural networks (conv. and recurrent) are more capable of extracting important features for the given task than classic ML approaches.
In this context, our team was able to produce further publications. {% cite elsner2019deep illium2020surgical muller2020analysis illium2022empirical%}
<center>
<figure class="half" style="max-width: 70%; text-align:center;">
<img src="\assets\images\projects\pipe_leak\5ea8f0ee-6c61-4835-944c-b77683d397ca.jpeg" style="margin-bottom: 0em;">
<img src="\assets\images\projects\pipe_leak\cc01cc58-d3f6-4220-b4f4-c7ea26b3a116.jpeg" style="margin-bottom: 0em;">
</figure>
</center><br>
This project ran from late 2018 until the early months of 2020.
{% include reference.html %}