Defect detection in reinforced concrete using random neural architectures

J.B. Butcher, C.R. Day, J.C. Austin, P.W. Haycock, D. Verstraeten, B. Schrauwen

Research output: Contribution to journalArticle

Abstract

Detecting defects within reinforced concrete is vital to the safety and durability of our built infrastructure upon which we heavily rely. In this work a non‐invasive technique, ElectroMagnetic Anomaly Detection (EMAD), is used which provides information into the electromagnetic properties of the reinforcing steel and for which data analysis is currently performed visually: an undesirable process. This article investigates the first use of two neural network approaches to automate the analysis of this data: Echo State Networks (ESNs) and Extreme Learning Machines (ELMs) where fast and efficient training procedures allow networks to be trained and evaluated in less time than traditional neural network approaches. Data collected from real‐world concrete structures have been analyzed using these two approaches as well as using a simple threshold measure and a standard recurrent neural network. The ELM approach offers a significant improvement in performance for a single tendon‐reinforced structure, while two ESN architectures provided best performance for a mesh‐reinforced concrete structure.
Original languageEnglish
Pages (from-to)191-207
JournalComputer-Aided Civil and Infrastructure Engineering
Volume29
Issue number3
DOIs
Publication statusPublished - 1 Mar 2014

Fingerprint

Concrete construction
Reinforced concrete
Learning systems
Neural networks
Recurrent neural networks
Network architecture
Durability
Defects
Steel
Defect detection

Cite this

Butcher, J. B., Day, C. R., Austin, J. C., Haycock, P. W., Verstraeten, D., & Schrauwen, B. (2014). Defect detection in reinforced concrete using random neural architectures. Computer-Aided Civil and Infrastructure Engineering, 29(3), 191-207. https://doi.org/10.1111/mice.12039
Butcher, J.B. ; Day, C.R. ; Austin, J.C. ; Haycock, P.W. ; Verstraeten, D. ; Schrauwen, B. / Defect detection in reinforced concrete using random neural architectures. In: Computer-Aided Civil and Infrastructure Engineering. 2014 ; Vol. 29, No. 3. pp. 191-207.
@article{ee425362b49f41718b1b75135c77e64c,
title = "Defect detection in reinforced concrete using random neural architectures",
abstract = "Detecting defects within reinforced concrete is vital to the safety and durability of our built infrastructure upon which we heavily rely. In this work a non‐invasive technique, ElectroMagnetic Anomaly Detection (EMAD), is used which provides information into the electromagnetic properties of the reinforcing steel and for which data analysis is currently performed visually: an undesirable process. This article investigates the first use of two neural network approaches to automate the analysis of this data: Echo State Networks (ESNs) and Extreme Learning Machines (ELMs) where fast and efficient training procedures allow networks to be trained and evaluated in less time than traditional neural network approaches. Data collected from real‐world concrete structures have been analyzed using these two approaches as well as using a simple threshold measure and a standard recurrent neural network. The ELM approach offers a significant improvement in performance for a single tendon‐reinforced structure, while two ESN architectures provided best performance for a mesh‐reinforced concrete structure.",
author = "J.B. Butcher and C.R. Day and J.C. Austin and P.W. Haycock and D. Verstraeten and B. Schrauwen",
year = "2014",
month = "3",
day = "1",
doi = "10.1111/mice.12039",
language = "English",
volume = "29",
pages = "191--207",
number = "3",

}

Butcher, JB, Day, CR, Austin, JC, Haycock, PW, Verstraeten, D & Schrauwen, B 2014, 'Defect detection in reinforced concrete using random neural architectures', Computer-Aided Civil and Infrastructure Engineering, vol. 29, no. 3, pp. 191-207. https://doi.org/10.1111/mice.12039

Defect detection in reinforced concrete using random neural architectures. / Butcher, J.B.; Day, C.R.; Austin, J.C.; Haycock, P.W.; Verstraeten, D.; Schrauwen, B.

In: Computer-Aided Civil and Infrastructure Engineering, Vol. 29, No. 3, 01.03.2014, p. 191-207.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Defect detection in reinforced concrete using random neural architectures

AU - Butcher, J.B.

AU - Day, C.R.

AU - Austin, J.C.

AU - Haycock, P.W.

AU - Verstraeten, D.

AU - Schrauwen, B.

PY - 2014/3/1

Y1 - 2014/3/1

N2 - Detecting defects within reinforced concrete is vital to the safety and durability of our built infrastructure upon which we heavily rely. In this work a non‐invasive technique, ElectroMagnetic Anomaly Detection (EMAD), is used which provides information into the electromagnetic properties of the reinforcing steel and for which data analysis is currently performed visually: an undesirable process. This article investigates the first use of two neural network approaches to automate the analysis of this data: Echo State Networks (ESNs) and Extreme Learning Machines (ELMs) where fast and efficient training procedures allow networks to be trained and evaluated in less time than traditional neural network approaches. Data collected from real‐world concrete structures have been analyzed using these two approaches as well as using a simple threshold measure and a standard recurrent neural network. The ELM approach offers a significant improvement in performance for a single tendon‐reinforced structure, while two ESN architectures provided best performance for a mesh‐reinforced concrete structure.

AB - Detecting defects within reinforced concrete is vital to the safety and durability of our built infrastructure upon which we heavily rely. In this work a non‐invasive technique, ElectroMagnetic Anomaly Detection (EMAD), is used which provides information into the electromagnetic properties of the reinforcing steel and for which data analysis is currently performed visually: an undesirable process. This article investigates the first use of two neural network approaches to automate the analysis of this data: Echo State Networks (ESNs) and Extreme Learning Machines (ELMs) where fast and efficient training procedures allow networks to be trained and evaluated in less time than traditional neural network approaches. Data collected from real‐world concrete structures have been analyzed using these two approaches as well as using a simple threshold measure and a standard recurrent neural network. The ELM approach offers a significant improvement in performance for a single tendon‐reinforced structure, while two ESN architectures provided best performance for a mesh‐reinforced concrete structure.

UR - http://www.scopus.com/inward/record.url?eid=2-s2.0-84892818956&partnerID=MN8TOARS

UR - https://onlinelibrary.wiley.com/doi/full/10.1111/mice.12039

U2 - 10.1111/mice.12039

DO - 10.1111/mice.12039

M3 - Article

VL - 29

SP - 191

EP - 207

IS - 3

ER -

Butcher JB, Day CR, Austin JC, Haycock PW, Verstraeten D, Schrauwen B. Defect detection in reinforced concrete using random neural architectures. Computer-Aided Civil and Infrastructure Engineering. 2014 Mar 1;29(3):191-207. https://doi.org/10.1111/mice.12039