Der Glasfaser-Internetanbieter gibt an, dass er mithilfe der vorhandenen Infrastruktur undichte Wasserleitungen erkennen kann und über einen Zeitraum von drei Monaten einen Verlust von 2 Millionen Litern pro Tag verhindert hat – die Lightsonic-Technologie erkennt unterirdische Vibrationen, maschinelles Lernen isoliert die Quelle
https://www.tomshardware.com/networking/fiber-provider-says-it-can-detect-leaking-water-pipes-using-existing-infrastructure-prevents-loss-of-2-million-liters-a-day-over-three-months-company-uses-lightsonic-technology-to-detect-underground-vibrations-machine-learning-to-isolate-source
2 Kommentare
>*“According to the company, the BT Group subsidiary is using Distributed Acoustic Sensing (DAS), a technology developed by Lightsonic, a startup firm developing monitoring technologies using existing fiber infrastructure. DAS analyzes changes in the light beam traveling through a fiber optic cable caused by underground vibrations.“*
The pessimist in me feels like this is an excuse to try to „legally“ read/scrape data from fiber lines. Water companies *already* use a combination of acoustic, pressure, and flow meters that allow them to natively narrow down the source of a leak to a block or less. This can be narrowed down to a few meters or less with a single field tech and an acoustic probe (if the location of the leak is not already obvious from the water coming out of the ground).
I’m just suspicious of any company promising cheap solutions to a problems as long as are allowed to use AI to analyze private data. We can detect water leaks bro, just use run the data streams through an algorithm bro.
This is the same concept behind using wifi signals to 3D scan the interior of buildings and obtain realtime position info for the people inside. IIRC they’ve been doing this for 6 or 7 years now.