Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional Network
In the practical application of the Bridge Weigh-In-Motion (BWIM) methods, the position of the wheels or axles during the passage of a vehicle is a prerequisite in most cases. To avoid the use of conventional axle detectors and bridge type-specific methods, we propose a novel method for axle detecti...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/22/8963 |
| _version_ | 1797463960487723008 |
|---|---|
| author | Steven Robert Lorenzen Henrik Riedel Maximilian Michael Rupp Leon Schmeiser Hagen Berthold Andrei Firus Jens Schneider |
| author_facet | Steven Robert Lorenzen Henrik Riedel Maximilian Michael Rupp Leon Schmeiser Hagen Berthold Andrei Firus Jens Schneider |
| author_sort | Steven Robert Lorenzen |
| collection | DOAJ |
| description | In the practical application of the Bridge Weigh-In-Motion (BWIM) methods, the position of the wheels or axles during the passage of a vehicle is a prerequisite in most cases. To avoid the use of conventional axle detectors and bridge type-specific methods, we propose a novel method for axle detection using accelerometers placed arbitrarily on a bridge. In order to develop a model that is as simple and comprehensible as possible, the axle detection task is implemented as a binary classification problem instead of a regression problem. The model is implemented as a Fully Convolutional Network to process signals in the form of Continuous Wavelet Transforms. This allows passages of any length to be processed in a single step with maximum efficiency while utilising multiple scales in a single evaluation. This allows our method to use acceleration signals from any location on the bridge structure and act as Virtual Axle Detectors (VADs) without being limited to specific structural types of bridges. To test the proposed method, we analysed 3787 train passages recorded on a steel trough railway bridge of a long-distance traffic line. Results of the measurement data show that our model detects 95% of the axles, which means that 128,599 out of 134,800 previously unseen axles were correctly detected. In total, 90% of the axles were detected with a maximum spatial error of 20 cm, at a maximum velocity of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>v</mi><mi>max</mi></msub><mo>=</mo><mn>56.3</mn><mspace width="3.33333pt"></mspace><mi mathvariant="normal">m</mi><mo>/</mo><mi mathvariant="normal">s</mi></mrow></semantics></math></inline-formula>. The analysis shows that our developed model can use accelerometers as VADs even under real operating conditions. |
| first_indexed | 2024-03-09T18:00:06Z |
| format | Article |
| id | doaj.art-7735123d71214d10a92a366b4d3dbd7c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T18:00:06Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-7735123d71214d10a92a366b4d3dbd7c2023-11-24T09:58:37ZengMDPI AGSensors1424-82202022-11-012222896310.3390/s22228963Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional NetworkSteven Robert Lorenzen0Henrik Riedel1Maximilian Michael Rupp2Leon Schmeiser3Hagen Berthold4Andrei Firus5Jens Schneider6Institute for Structural Mechanics and Design, Technical University of Darmstadt, 64287 Darmstadt, GermanyInstitute for Structural Mechanics and Design, Technical University of Darmstadt, 64287 Darmstadt, GermanyInstitute for Structural Mechanics and Design, Technical University of Darmstadt, 64287 Darmstadt, GermanyInstitute for Structural Mechanics and Design, Technical University of Darmstadt, 64287 Darmstadt, GermanyInstitute for Structural Mechanics and Design, Technical University of Darmstadt, 64287 Darmstadt, GermanyiSEA Tec GmbH, 88046 Friedrichshafen, GermanyInstitute for Structural Mechanics and Design, Technical University of Darmstadt, 64287 Darmstadt, GermanyIn the practical application of the Bridge Weigh-In-Motion (BWIM) methods, the position of the wheels or axles during the passage of a vehicle is a prerequisite in most cases. To avoid the use of conventional axle detectors and bridge type-specific methods, we propose a novel method for axle detection using accelerometers placed arbitrarily on a bridge. In order to develop a model that is as simple and comprehensible as possible, the axle detection task is implemented as a binary classification problem instead of a regression problem. The model is implemented as a Fully Convolutional Network to process signals in the form of Continuous Wavelet Transforms. This allows passages of any length to be processed in a single step with maximum efficiency while utilising multiple scales in a single evaluation. This allows our method to use acceleration signals from any location on the bridge structure and act as Virtual Axle Detectors (VADs) without being limited to specific structural types of bridges. To test the proposed method, we analysed 3787 train passages recorded on a steel trough railway bridge of a long-distance traffic line. Results of the measurement data show that our model detects 95% of the axles, which means that 128,599 out of 134,800 previously unseen axles were correctly detected. In total, 90% of the axles were detected with a maximum spatial error of 20 cm, at a maximum velocity of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mi>v</mi><mi>max</mi></msub><mo>=</mo><mn>56.3</mn><mspace width="3.33333pt"></mspace><mi mathvariant="normal">m</mi><mo>/</mo><mi mathvariant="normal">s</mi></mrow></semantics></math></inline-formula>. The analysis shows that our developed model can use accelerometers as VADs even under real operating conditions.https://www.mdpi.com/1424-8220/22/22/8963moving load localisationnothing-on-roadfree-of-axle-detectorbridge weigh-in-motionstructural health monitoringfield validation |
| spellingShingle | Steven Robert Lorenzen Henrik Riedel Maximilian Michael Rupp Leon Schmeiser Hagen Berthold Andrei Firus Jens Schneider Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional Network Sensors moving load localisation nothing-on-road free-of-axle-detector bridge weigh-in-motion structural health monitoring field validation |
| title | Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional Network |
| title_full | Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional Network |
| title_fullStr | Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional Network |
| title_full_unstemmed | Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional Network |
| title_short | Virtual Axle Detector Based on Analysis of Bridge Acceleration Measurements by Fully Convolutional Network |
| title_sort | virtual axle detector based on analysis of bridge acceleration measurements by fully convolutional network |
| topic | moving load localisation nothing-on-road free-of-axle-detector bridge weigh-in-motion structural health monitoring field validation |
| url | https://www.mdpi.com/1424-8220/22/22/8963 |
| work_keys_str_mv | AT stevenrobertlorenzen virtualaxledetectorbasedonanalysisofbridgeaccelerationmeasurementsbyfullyconvolutionalnetwork AT henrikriedel virtualaxledetectorbasedonanalysisofbridgeaccelerationmeasurementsbyfullyconvolutionalnetwork AT maximilianmichaelrupp virtualaxledetectorbasedonanalysisofbridgeaccelerationmeasurementsbyfullyconvolutionalnetwork AT leonschmeiser virtualaxledetectorbasedonanalysisofbridgeaccelerationmeasurementsbyfullyconvolutionalnetwork AT hagenberthold virtualaxledetectorbasedonanalysisofbridgeaccelerationmeasurementsbyfullyconvolutionalnetwork AT andreifirus virtualaxledetectorbasedonanalysisofbridgeaccelerationmeasurementsbyfullyconvolutionalnetwork AT jensschneider virtualaxledetectorbasedonanalysisofbridgeaccelerationmeasurementsbyfullyconvolutionalnetwork |