2018
DOI: 10.15446/dyna.v85n204.68408
|View full text |Cite
|
Sign up to set email alerts
|

Fast estimation of earthquake epicenter distance using a single seismological station with machine learning techniques

Abstract: A Support Vector Machine Regression (SVMR) algorithm was applied to calculate the epicenter distance using a ten seconds signal, after primary waves arrive at a seismological station near to Bogota -Colombia. This algorithm was tested with 863 records of earthquakes, where the input parameters were an exponential function of waveform envelope estimated by least squares and maximum value of recorded waveforms for each component of the seismic station. Cross validation was applied to normalized polynomial kernel… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
3
0
2

Year Published

2019
2019
2024
2024

Publication Types

Select...
9

Relationship

3
6

Authors

Journals

citations
Cited by 12 publications
(5 citation statements)
references
References 16 publications
0
3
0
2
Order By: Relevance
“…Recently, Machine Learning (ML)-based approaches to localization have also seen a wider use. While some are based on waveforms detected at different stations or even single stations [31,32], others make use of the wave arrival times to estimate the source of the earthquake in a similar way to some of the previously described methods. Convolutional neural networks [33], attention-based networks [34], and other ML methods [35] have shown good accuracy in determining the epicentral/hypocentral position, with different degrees of spatial resolution; however, to the authors' knowledge, most have either been tested in a vacuum or do not provide figures on their time performance, with the exception of [35], which provides some estimated figures on its implementation in an EW scenario.…”
Section: Related Workmentioning
confidence: 99%
“…Recently, Machine Learning (ML)-based approaches to localization have also seen a wider use. While some are based on waveforms detected at different stations or even single stations [31,32], others make use of the wave arrival times to estimate the source of the earthquake in a similar way to some of the previously described methods. Convolutional neural networks [33], attention-based networks [34], and other ML methods [35] have shown good accuracy in determining the epicentral/hypocentral position, with different degrees of spatial resolution; however, to the authors' knowledge, most have either been tested in a vacuum or do not provide figures on their time performance, with the exception of [35], which provides some estimated figures on its implementation in an EW scenario.…”
Section: Related Workmentioning
confidence: 99%
“…Grilla de búsqueda del error medio absoluto, con el cual, es posible medir directamente cuanto se desvía la predicción. El error encontrado con los parámetros seleccionados, quiere decir que los valores predichos se estimaron con 0,22 barns/ electrón de error absoluto en el set de entrenamiento (Ochoa et al, 2018). Lo cual puede ser una aproximación, ya que esta tabla no contiene una tendencia central, a diferencia de las otras dos grillas que si la presentan.…”
Section: Machine Learning Con Wekaunclassified
“…La función clasificadora utilizada es SMOreg con un Normalized Polykernel. Esta técnica también se utiliza para el resto de las predicciones, y se configuró así teniendo en cuenta la asesoría del autor que realizó la estimación de un epicentro utilizando el mismo programa (Ochoa et al, 2018). Los resultados obtenidos para el set de entrenamiento se muestran en la Figura 8.…”
Section: Error Medio Cuadráticounclassified
“…There are several methods to detect seismic wave and its arrival azimuth in a single three-component station (Magotra et al, 1987;Anant & Dowla, 1997); these authors employed algorithms that measure the level of linear polarization in the P wave's arrival. The methodology proposed in this research consists of applying SVMs along with a kernel function in order to estimate the arrival azimuth with minimal processing of data acquired at the station, similar to methodology applied to a fast determination of earthquake magnitude and epicenter distance using a single seismological station (Ochoa et al, 2017;Ochoa et al, 2018).…”
Section: Introductionmentioning
confidence: 99%