Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor Data

Sinkholes are geohazards that commonly form in karstifiable terrain and are an ever-present danger to infrastructure and human life. This paper aims to answer the question: Can a cover–subsidence sinkhole’s volume be determined using fibre Bragg grating sensor strain data and machine-learning techni...

Full description

Saved in:
Bibliographic Details
Main Authors: Wesley B. Richardson, Suné von Solms, Johan Meyer, Charis Harley
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/7/2272
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849769347350790144
author Wesley B. Richardson
Suné von Solms
Johan Meyer
Charis Harley
author_facet Wesley B. Richardson
Suné von Solms
Johan Meyer
Charis Harley
author_sort Wesley B. Richardson
collection DOAJ
description Sinkholes are geohazards that commonly form in karstifiable terrain and are an ever-present danger to infrastructure and human life. This paper aims to answer the question: Can a cover–subsidence sinkhole’s volume be determined using fibre Bragg grating sensor strain data and machine-learning techniques? Exploratory data analysis was conducted on fibre Bragg grating sensor strain data collected from an experimental test rig whereby a cover–subsidence sinkhole was formed. It was found that statistical techniques and machine-learning algorithms that assume normality are inappropriate when performing phase classification and volume regression tasks on the cover–subsidence sinkhole when given fibre Bragg grating sensor’s strain data. Weighted Least Squares regression, Support Vector Regression, and eXtreme Gradient Boosting were implemented on the data during phase two of the cover–subsidence sinkhole formation to determine the volume of the sinkhole. Weighted Least Squares regression obtained the lowest <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula> values for training and testing. Support Vector Regression had significantly improved results over Weighted Least Squares regression, while eXtreme Gradient Boosting obtained the highest <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula> values for training and testing. The highest <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula> values for eXtreme Gradient Boosting obtained were <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>1.00</mn></mrow></semantics></math></inline-formula> for training and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.97</mn></mrow></semantics></math></inline-formula> for testing. In addition, eXtreme Gradient Boosting had the lowest root mean squared errors compared to Weighted Least Squares regression and Support Vector Regression. It was found that eXtreme Gradient Boosting is a strong candidate for determining the volume of the C–S sinkhole when using fibre Bragg grating strain data.
format Article
id doaj-art-e7bc750390964329abd504970d1d9725
institution DOAJ
issn 1424-8220
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-e7bc750390964329abd504970d1d97252025-08-20T03:03:27ZengMDPI AGSensors1424-82202025-04-01257227210.3390/s25072272Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor DataWesley B. Richardson0Suné von Solms1Johan Meyer2Charis Harley3Department of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South AfricaDepartment of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South AfricaDepartment of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South AfricaDepartment of Electrical and Electronic Engineering Science, University of Johannesburg, Johannesburg 2006, South AfricaSinkholes are geohazards that commonly form in karstifiable terrain and are an ever-present danger to infrastructure and human life. This paper aims to answer the question: Can a cover–subsidence sinkhole’s volume be determined using fibre Bragg grating sensor strain data and machine-learning techniques? Exploratory data analysis was conducted on fibre Bragg grating sensor strain data collected from an experimental test rig whereby a cover–subsidence sinkhole was formed. It was found that statistical techniques and machine-learning algorithms that assume normality are inappropriate when performing phase classification and volume regression tasks on the cover–subsidence sinkhole when given fibre Bragg grating sensor’s strain data. Weighted Least Squares regression, Support Vector Regression, and eXtreme Gradient Boosting were implemented on the data during phase two of the cover–subsidence sinkhole formation to determine the volume of the sinkhole. Weighted Least Squares regression obtained the lowest <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula> values for training and testing. Support Vector Regression had significantly improved results over Weighted Least Squares regression, while eXtreme Gradient Boosting obtained the highest <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula> values for training and testing. The highest <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula> values for eXtreme Gradient Boosting obtained were <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>1.00</mn></mrow></semantics></math></inline-formula> for training and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.97</mn></mrow></semantics></math></inline-formula> for testing. In addition, eXtreme Gradient Boosting had the lowest root mean squared errors compared to Weighted Least Squares regression and Support Vector Regression. It was found that eXtreme Gradient Boosting is a strong candidate for determining the volume of the C–S sinkhole when using fibre Bragg grating strain data.https://www.mdpi.com/1424-8220/25/7/2272cover–subsidence sinkholeexploratory data analysisextreme gradient boostfibre bragg gratingmachine learningstrain data
spellingShingle Wesley B. Richardson
Suné von Solms
Johan Meyer
Charis Harley
Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor Data
Sensors
cover–subsidence sinkhole
exploratory data analysis
extreme gradient boost
fibre bragg grating
machine learning
strain data
title Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor Data
title_full Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor Data
title_fullStr Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor Data
title_full_unstemmed Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor Data
title_short Prediction of Cover–Subsidence Sinkhole Volume Using Fibre Bragg Grating Strain Sensor Data
title_sort prediction of cover subsidence sinkhole volume using fibre bragg grating strain sensor data
topic cover–subsidence sinkhole
exploratory data analysis
extreme gradient boost
fibre bragg grating
machine learning
strain data
url https://www.mdpi.com/1424-8220/25/7/2272
work_keys_str_mv AT wesleybrichardson predictionofcoversubsidencesinkholevolumeusingfibrebragggratingstrainsensordata
AT sunevonsolms predictionofcoversubsidencesinkholevolumeusingfibrebragggratingstrainsensordata
AT johanmeyer predictionofcoversubsidencesinkholevolumeusingfibrebragggratingstrainsensordata
AT charisharley predictionofcoversubsidencesinkholevolumeusingfibrebragggratingstrainsensordata