Estimating Fault Friction From Seismic Signals in the Laboratory

2019

The seismogenic plate boundaries are presumed to behave similarly to a densely packed granular medium, where fault and blocks systems rapidly rearrange the distribution of forces within themselves, as particles do in slowly sheared granular systems. We use machine learning and show that statistical features of velocity signals from individual particles in a simulated sheared granular fault contain information regarding the instantaneous global state of intermittent frictional stick-slip dynamics. We demonstrate that combining features built from the signals of more particles can improve the accuracy of the global model, and discuss the physical basis behind decrease in error. We show that the statistical features such as median and higher moments of the signals that represent the particle displacement in the direction of shearing are among the best predictive features. Our work provides novel insights into the applications of machine learning in studying frictional processes that ta...

arXiv: Geophysics, 2018

Over the last two decades, strain and GPS measurements have shown that slow slip on earthquake faults is a widespread phenomenon. Slow slip is also inferred from correlated small amplitude seismic signals known as nonvolcanic tremor and low frequency earthquakes (LFEs). Slow slip has been reproduced in laboratory and simulation studies, however the fundamental physics of these phenomena and their relationship to dynamic earthquake rupture remains poorly understood. Here we show that, in a laboratory setting, continuous seismic waves are imprinted with fundamental signatures of the fault's physical state. Using machine learning on continuous seismic waves, we can infer several bulk characteristics of the fault (friction, shear displacement, gouge thickness), at any time during the slow slip cycle. This analysis also allows us to infer many properties of the future behavior of the fault, including the time remaining before the next slow slip event. Our work suggests that by applyi...

Geophysical Research Letters, 2018

Machine learning regression can predict macroscopic fault properties such as shear stress, friction, and time to failure using continuous records of fault zone acoustic emissions. Here we show that a similar approach is successful using event catalogs derived from the continuous data. Our methods are applicable to catalogs of arbitrary scale and magnitude of completeness. We investigate how machine learning regression from an event catalog of laboratory earthquakes performs as a function of the catalog magnitude of completeness. We find that strong model performance requires a sufficiently low magnitude of completeness, and below this magnitude of completeness, model performance saturates.

Proceedings, 2014

2019

In this paper we find a pattern of aperiodic seismic signals that precede earthquakes at any time in a laboratory earthquake’s cycle using a small window of time. We use a data set that comes from a classic laboratory experiment having several stick-slip displacements (earthquakes), a type of experiment which has been studied as a simulation of seismologic faults for decades. This data exhibits similar behavior to natural earthquakes, so the same approach may work in predicting the timing of them. Here we show that by applying random forest machine learning technique to the acoustic signal emitted by a laboratory fault, we can predict the time remaining before failure with 1.61 seconds mean absolute error at any moment of earthquake’s cycle. These predictions are based solely on the acoustical signal’s statistical features derived from the local, moving 0.3 second time windows and do not make use of its history. Essential improvements in providing new understanding of fault physics ...

arXiv: Geophysics, 2020

The successful prediction of earthquakes is one of the holy grails in Earth Sciences. Traditional predictions use statistical information on recurrence intervals, but those predictions are not accurate enough. In a recent paper, a machine learning approach was proposed and applied to data of laboratory earthquakes. The machine learning algorithm utilizes continuous measurements of radiated energy through acoustic emissions and the authors were able to successfully predict the timing of laboratory earthquakes. Here, we reproduced their model which was applied to a gouge layer of glass beads and applied it to a data set obtained using a gouge layer of salt. In this salt experiment different load point velocities were set, leading to variable recurrence times. The machine learning technique we use is called random forest and uses the acoustic emissions during the interseismic period. The random forest model succeeds in making a relatively reliable prediction for both materials, also lo...

Geophysical Research Letters

2019

*The paper was submitted as an Extended Essay to the International Baccalaureate Organization in 2019 and earned the highest grade, A. This work explores classic regression and classification algorithms, linear and logistic, as well as the relatively new model Support Vector Machines to determine how the differences in the mathematical formalism of machine learning algorithms affect the accuracy of regression and classification. A dataset containing records of seismic activity in Pakistan was then chosen and the software Weka was used to compare the accuracy of regression and classification of both machine learning algorithms. It was found that they produced similar results.

Sensors

This article aims to discusses machine learning modelling using a dataset provided by the LANL (Los Alamos National Laboratory) earthquake prediction competition hosted by Kaggle. The data were obtained from a laboratory stick-slip friction experiment that mimics real earthquakes. Digitized acoustic signals were recorded against time to failure of a granular layer compressed between steel plates. In this work, machine learning was employed to develop models that could predict earthquakes. The aim is to highlight the importance and potential applicability of machine learning in seismology The XGBoost algorithm was used for modelling combined with 6-fold cross-validation and the mean absolute error (MAE) metric for model quality estimation. The backward feature elimination technique was used followed by the forward feature construction approach to find the best combination of features. The advantage of this feature engineering method is that it enables the best subset to be found from...

International Journal on Recent and Innovation Trends in Computing and Communication, 2023

The most common causes of coal mining accidents are seismic hazard, fires, explosions, and landslips. These accidents are usually caused by various factors such as mechanical and technical failures, as well as social and economic factors. An analysis of these accidents can help identify the exact causes of these accidents and prevent them from happening in the future. There are also various seismic events that can occur in underground mines. These include rock bumps and tremors. These have been reported in different countries such as Australia, China, France, Germany, India, Russia, and Poland. Through the use of advanced seismological and seismic monitoring systems, we can now better understand the rock mass processes that can cause a seismic hazard. Unfortunately, despite the advancements, the accuracy of these methods is still not perfect. One of the main factors that prevent the development of effective seismic hazard prediction techniques is the complexity of the seismic processes. In order to carry out effective seismic risk assessment in mines, it is important that the discrimination of seismicity in different regions is carried out. The widespread use of machine learning in analyzing seismic data, it provides reliability and feasibility for preventing major mishaps. This paper provides uses various machine learning classifiers to predict seismic hazards