Integration of Machine Learning on Distributed Acoustic Sensing Surveys

2020 ◽  
Author(s):  
Camille Jestin ◽  
Clément Hibert ◽  
Gaëtan Calbris ◽  
Vincent Lanticq
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Fibre Optic ◽  
Rock Falls ◽  
Near Surface ◽  
Acoustic Sensing ◽  
Seismic Waveforms ◽  
Fibre Optic Cable ◽  
Distributed Acoustic Sensing

<p>Distributed Acoustic Sensing (DAS) is an innovative technique which has been recently employed for near-surface geophysics purposes. It involves the use of fibre-optic cable as a sensor. The fibre is analysed by sending a laser pulse from an interrogator unit. The phase of the backscattered signal contains the information on the strain on the cable, enabling the detection of a passing acoustic wave with enough energy for the cable excitation. Allowing the interrogation of long profiles and the generation of a dense spatial sampling, uneasy to obtain with classic geophysical techniques, DAS instrumentation then proved its relevance for seismic applications but also for infrastructure monitoring.</p><p>During DAS acquisition, and more precisely when closely looking at infrastructures integrity, it is necessary to clearly identify the source of the acoustic vibrations at the structure neighbourhood. Indeed, in the context of pipeline monitoring for example, it appears important to be able to classify events which generate seismic signals recorded by DAS systems and which can be related to a potential threat for the structure. In order to launch an alarm if necessary, the source identification must be fast, accurate and robust. Moreover, because DAS acquisition can generate traces every few meters along fibres of tens of kilometres, the used machine-learning algorithm must demonstrate its ability to handle a big amount of data.</p><p>In this study, we analyse the efficiency of the Random Forests (RF) machine-learning algorithm applied to data acquired with DAS system for the discrimination of event sources. RF algorithm has been selected because of its ability to handle large numbers of attributes related to signal characteristics and to enable a good reliability for the discrimination of sources. This algorithm has already proved its efficiency for automated classification of seismic waveforms (e.g. earthquakes, volcanic tremors, rock falls, avalanches, etc.).</p><p>We focus our study on tests lead along a gas pipeline instrumented with fibre-optic cable. Different third-party works have been conducted: excavation, saw sections, drill, jackhammer, etc. We work on the discrimination of six classes of seismic source. After running a detection phase based on a threshold on signal energy, we obtain several hundred of exploitable seismic traces to inject to the RF algorithm. We demonstrate the efficiency of the application of machine learning on DAS data to discriminate seismic waveforms from the correct class, with an overall precision on our test set of 99%.</p>

Danish earthquake recorded by distributed acoustic sensing (DAS)

2020 ◽  
Author(s):  
Camilla Rasmussen ◽  
Peter H. Voss ◽  
Trine Dahl-Jensen
Keyword(s):  
Field Test ◽  
Signal To Noise Ratio ◽  
Field Tests ◽  
Small Subset ◽  
Fibre Optic ◽  
Data Set ◽  
Acoustic Sensing ◽  
Arrival Times ◽  
Fibre Optic Cable ◽  
Distributed Acoustic Sensing

<p>On September 16th 2018 a Danish earthquake of local magnitude 3.7 was recorded by distributed acoustic sensing (DAS) in a ~23 km long fibre-optic cable. The data are used to study how well DAS can be used as a supplement to conventional seismological data in earthquake localisation. One of the goals in this study is extracting a small subset of traces with clear P and S phases to use in an earthquake localisation, from the 11144 traces the DAS system provide. The timing in the DAS data might not be reliable, and therefore differences in arrival times of S and P are used instead of the exact arrival times. <br>The DAS data set is generally noisy and with a low signal-to-noise ratio (SNR). It is examined whether stacking can be used to improve SNR. The SNR varies a lot along the fibre-optic cable, and at some distances, it is so small that the traces are useless. Stacking methods for improving SNR are presented.</p><p>A field test at two location sites of the fibre-optic cable was conducted with the purpose of comparing DAS data with seismometer data. At the field sites, hammer shots were recorded by a small array of three STS-2 sensors located in a line parallel to the fibre-optic cable. The recordings generally show good consistency between the two data sets. <br>In addition, the field tests are used to get a better understanding of the noise sources in the DAS recording of the earthquake. There are many sources of noise in the data set. The most prominent are a line of windmills that cross the fibre-optic cable and people walking in the building where the detector is located. Also, the coupling between the fibre-optic cable and the ground varies along the cable length due to varying soil type and wrapping around the fibre-optic cable, which is also evident in field test data. Furthermore, the data from the field tests are used to calibrate the location of the fibre-optic cable, which is necessary for using the DAS data in an earthquake localisation. <br>Data processing is done in Matlab and SEISAN.</p>

Observing avalanche dynamics with Distributed Acoustic Sensing

2021 ◽  
Author(s):  
Andreas Fichtner ◽  
Pascal Edme ◽  
Patrick Paitz ◽  
Nadja Lindner ◽  
Michael Hohl ◽  
...  
Keyword(s):  
Spatial Resolution ◽  
Flow Regimes ◽  
Dynamic Strain ◽  
Seismic Signals ◽  
Fibre Optic ◽  
Snow Avalanches ◽  
Acoustic Sensing ◽  
Fibre Optic Cable ◽  
Distributed Acoustic Sensing ◽  
Seismic Instrumentation

<p><span>Avalanche research requires comprehensive measurements of sudden and rapid snow mass movement that is hard to predict. Automatic cameras, radar and infrasound sensors provide valuable observations of avalanche structure and dynamic parameters, such as velocity. Recently, seismic sensors have also gained popularity, because they can monitor avalanche activity over larger spatial scales. Moreover, seismic signals elucidate rheological properties, which can be used to distinguish different types of avalanches and flow regimes. To date, however, seismic instrumentation in avalanche terrain is sparse. This limits the spatial resolution of avalanche details, needed to characterise flow regimes and maximise detection accuracy for avalanche warning.</span></p><p><span>As an alternative to conventional seismic instrumentation, we propose Distributed Acoustic Sensing (DAS) to measure avalanche-induced ground motion. DAS is based on fibre-optic technology, which has previously been used already for environmental monitoring, e.g., of snow avalanches. Thanks to recent technological advances, modern DAS interrogators allow us to measure dynamic strain along a fibre-optic cable with unprecedented temporal and spatial resolution. It therefore becomes possible to record seismic signals along many kilometres of fibre-optic cables, with a spatial resolution of a few metres, thereby creating large arrays of seismic receivers. We test this approach at an avalanche test site in the Valleé de la Sionne, in the Swiss Alps, using an existing 700 m long fibre-optic cable that is permanently installed underground for the purpose of data transfer of other, independent avalanche measurements.</span></p><p><span>During winter 2020/2021, we recorded numerous snow avalanches, including several which reached the valley bottom, travelling directly over the cable during runout. The DAS recordings show clear seismic signatures revealing individual flow surges and various phases/modes that may be associated with roll waves and avalanche arrest. We compare our observations to state-of-the-art radar and seismic measurements which ideally complement the DAS data.</span></p><p><span>Our initial analysis highlights the suitability of DAS-based monitoring and research for avalanches and other hazardous granular flows. Moreover, the clear detectability of avalanche signals using existing fibre-optic infrastructure of telecommunication networks opens the opportunity for unrivalled warning capabilities in Alpine environments.</span></p>

Fibre optic Distributed Acoustic Sensing of volcanic events at Mt Etna

2020 ◽  
Author(s):  
Gilda Currenti ◽  
Philippe Jousset ◽  
Athena Chalari ◽  
Luciano Zuccarello ◽  
Rosalba Napoli ◽  
...  
Keyword(s):  
Acoustic Wave ◽  
Seismic Waves ◽  
Dynamic Strain ◽  
Fibre Optic ◽  
Etna Volcano ◽  
Acoustic Sensing ◽  
Volcanic Events ◽  
Fibre Optic Cable ◽  
Distributed Acoustic Sensing ◽  
Explosive Event

<p>We explore a unique dataset collected by Distributed Acoustic Sensing (DAS) technology at the summit of Etna volcano in September 2018. We set-up an iDAS interrogator (Silixa) inside the Observatory Pizzi Deneri to record strain rate signals along a 1.3 km-long fibre optic cable deployed in Piano delle Concazze. This area is affected by several North-South trending faults and fractures, that are originated to accommodate the extension of the nearby North-East Rift zone, where magmatic intrusions often occur. The field evidence of the segments of these faults and fractures is hidden by lava flows and volcano-clastic deposits (e.g. scoria and lapilli) produced by the effusive and explosive activity of Etna volcano.</p><p>We propose a new technological and methodological framework to validate, identify and characterize volcano-related dynamic strain changes at an unprecedented high spatial (2 m) and temporal (1 kHz) sampling over a broad frequency range. The DAS record analysis and the validation of the iDAS response is performed through comparisons with measurements from a dense network of conventional sensors (comprising 5 broadband seismometers, 15 short-period geophone and two arrays of 3 infrasound sensors) deployed along  the fibre optic cable.  Comparisons between iDAS signals and dynamic strain changes estimated from the broadband seismic array shows an excellent agreement, thus demonstrating for the first time the capability of DAS technology in sensing seismic waves generated by volcanic events.</p><p>The frequent and diverse Etna activity during the acquisition period (30 August - 16 September 2018) offers the great opportunity to record a wide variety of signals and, hence, to test the response of iDAS to several volcanic processes (e.g. volcanic tremor, explosions, strombolian activity, local seismic events). Here, we focus the analysis on the signals recorded during a small explosive event on 5 September 2018 from the New South-East Crater (NSEC). This explosive event generated both seismic waves (recorded by the seismometers) propagating in the ground, and acoustic pressure signals (recorded by the infra-sound arrays) propagating in the atmosphere. We show that the DAS records catch both, as confirmed by the conventional sensors records.</p><p>Spectrogram analyses of the DAS signals reveal that the frequency content is confined in two distinctive frequency bands in the ranges 0.5-10 Hz and 18-25 Hz, for the seismic and acoustic wave, respectively. The amplitude and frequency response of the ground to the arrival and propagation of the seismo-acoustic wave along the fibre reveal spatial characteristic patterns that reflect local geological structures. For example, the finer spatial sampling of the iDAS records allows catching details of the variability of dynamic strain amplitudes along the fibre. Amplified signals are found at localized narrow regions matching fracture zones and faults. There, a decrease in the propagation velocity of the seismo-acoustic waves is also clearly pinpointed. </p><p>These preliminary findings demonstrate the DAS potentiality to revolutionize the study of volcanic process by discovering new signal features undetectable with traditional sensors and methodologies.</p>

Machine Learning Algorithm to Predict Early Complications after Brain Tumor Surgery

2018 ◽  
Author(s):  
C.H.B. van Niftrik ◽  
F. van der Wouden ◽  
V. Staartjes ◽  
J. Fierstra ◽  
M. Stienen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Brain Tumor ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Tumor Surgery ◽  
Early Complications ◽  
Brain Tumor Surgery

Feature extraction and prediction of Dengue Outbreaks

2020 ◽  
pp. 216-222
Author(s):  
Kunal Parikh ◽  
Tanvi Makadia ◽  
Harshil Patel
Keyword(s):  
Public Health ◽  
Machine Learning ◽  
Developing Countries ◽  
Feature Extraction ◽  
Predictive Analytics ◽  
Learning Algorithm ◽  
Machine Learning Algorithm ◽  
Health Concerns ◽  
The World ◽  
Dengue Outbreaks

Dengue is unquestionably one of the biggest health concerns in India and for many other developing countries. Unfortunately, many people have lost their lives because of it. Every year, approximately 390 million dengue infections occur around the world among which 500,000 people are seriously infected and 25,000 people have died annually. Many factors could cause dengue such as temperature, humidity, precipitation, inadequate public health, and many others. In this paper, we are proposing a method to perform predictive analytics on dengue’s dataset using KNN: a machine-learning algorithm. This analysis would help in the prediction of future cases and we could save the lives of many.

Sign in / Sign up

Export Citation Format

Share Document