Communications of the ACM
Machine Learning sed to Predict Earthquakes in a Lab Setting
Researchers say they have identified a hidden signal that leads up to earthquakes, and have used it to train a machine-learning algorithm to predict future quakes.
Credit: University of Cambridge
Researchers at the University of Cambridge in the U.K., Los Alamos National Laboratory, and Boston University say they have identified a hidden signal that leads up to earthquakes, and have used it to train a machine-learning algorithm to predict future seismic events.
The team set up a laboratory apparatus that uses steel blocks to mimic the physical forces at work in an actual earthquake while recording the seismic signals and sounds that are emitted. They then used machine learning to identify the relationship between the acoustic signal coming from the "fault" and how close it is to failing.
The algorithm identified a specific pattern in the sound, previously considered noise, which occurs before a quake hits.
The researchers note the characteristics of this sound pattern can be used to develop a precise estimate of the stress on the fault and to estimate the time remaining prior to failure.
From University of Cambridge
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found