Calibration of a Finite Element Forward Model in Eddy Current Inspection.

Joel Hampton; Henok Tesfalem; Oliver Dorn; Adam Fletcher; Anthony Peyton; Matthew Brown

doi:10.1109/JSEN.2022.3167253

Calibration of a Finite Element Forward Model in Eddy Current Inspection.

Joel Hampton
, Henok Tesfalem
, Oliver Dorn
, Adam Fletcher
, Anthony Peyton
, Matthew Brown

EDF Energy Nuclear Generation Ltd

Research output: Contribution to journal › Article › peer-review

384 Downloads (Pure)

Abstract

We report on the use of a novel constrained optimisation algorithm for calibrating the finite element model in an eddy current inspection application. An accurate finite element forward model is often important in such eddy current applications for training neural networks or as part of an iterative solver. However, the subject of calibration of the model has not received much attention in the literature to date.We consider amulti-frequency, eddy current depth profiling application, which is important for non-destructive testing in the nuclear industry. In the optimisation algorithm, we use a Levenberg-Marquardt algorithm and a bisection search to ensure constraints are satisfied, coupled with a truncated Gradient and Hessian method. We calibrate two types of finite element models, one using a filament representation for the coils and the other using a full 3D approach. The results show the feasibility of using the constrained non-linear optimisation algorithm for tuning the finite element model parameters. The mean signal-noise ratio after tuning on a truncated spectrum was 29.47 dB for the 3D model and 28.96 dB for the filament; in contrast the mean SNR using the measured coil parameters was 1.48 dB and 4.98 dB for the two uncalibrated models, respectively. The results show that a filament model is competitive with a 3D model of the coils (within the nuclear graphite application); therefore, a computationally faster filament model can be used with minimal effect on accuracy.

Original language	English
Journal	IEEE Sensors Journal
DOIs	https://doi.org/10.1109/JSEN.2022.3167253
Publication status	Published - 18 Apr 2022

Access to Document

10.1109/JSEN.2022.3167253

calibration_paperAccepted author manuscript, 4.77 MB

Electromagnetic Sensing Group
Peyton, A. (PI), Fletcher, A. (Researcher), Daniels, D. (CoI), Conniffe, D. (PGR student), Podd, F. (PI), Davidson, J. (Researcher), Anderson, J. (Support team), Wilson, J. (Researcher), Marsh, L. (PI), O'Toole, M. (PI), Watson, S. (PGR student), Yin, W. (PI), Regan, A. (PGR student), Williams, K. (Researcher), Rana, S. (Researcher), Khalil, K. (PGR student), Hills, D. (PGR student), Whyte, C. (PGR student), Wang, C. (PGR student), Hodgskin-Brown, R. (PGR student), Dadkhahtehrani, F. (PGR student), Forster, S. (PGR student), Zhu, F. (PGR student), Yu, K. (PGR student), Xiong, L. (PGR student), Lu, T. (PGR student), Zhang, L. (PGR student), Lyu, R. (PGR student), Zhu, R. (PGR student), She, S. (PGR student), Meng, T. (PGR student), Pang, X. (PGR student), Zheng, X. (PGR student), Bai, X. (PGR student), Zou, X. (PGR student), Ding, Y. (PGR student), Shao, Y. (PGR student), Xia, Z. (PGR student), Zhang, Z. (PGR student), Khangerey, R. (PGR student), Lawless, B. (Researcher), Lawless, B. (PI) & Khalil, K. (PI)
1/10/04 → …
Project: Research

Cite this

@article{76f8ac5210404ac99fd638dcbdefbf2d,

title = "Calibration of a Finite Element Forward Model in Eddy Current Inspection.",

abstract = "We report on the use of a novel constrained optimisation algorithm for calibrating the finite element model in an eddy current inspection application. An accurate finite element forward model is often important in such eddy current applications for training neural networks or as part of an iterative solver. However, the subject of calibration of the model has not received much attention in the literature to date.We consider amulti-frequency, eddy current depth profiling application, which is important for non-destructive testing in the nuclear industry. In the optimisation algorithm, we use a Levenberg-Marquardt algorithm and a bisection search to ensure constraints are satisfied, coupled with a truncated Gradient and Hessian method. We calibrate two types of finite element models, one using a filament representation for the coils and the other using a full 3D approach. The results show the feasibility of using the constrained non-linear optimisation algorithm for tuning the finite element model parameters. The mean signal-noise ratio after tuning on a truncated spectrum was 29.47 dB for the 3D model and 28.96 dB for the filament; in contrast the mean SNR using the measured coil parameters was 1.48 dB and 4.98 dB for the two uncalibrated models, respectively. The results show that a filament model is competitive with a 3D model of the coils (within the nuclear graphite application); therefore, a computationally faster filament model can be used with minimal effect on accuracy.",

author = "Joel Hampton and Henok Tesfalem and Oliver Dorn and Adam Fletcher and Anthony Peyton and Matthew Brown",

year = "2022",

month = apr,

day = "18",

doi = "10.1109/JSEN.2022.3167253",

language = "English",

journal = "IEEE Sensors Journal",

issn = "1530-437X",

publisher = "IEEE",

}

TY - JOUR

T1 - Calibration of a Finite Element Forward Model in Eddy Current Inspection.

AU - Hampton, Joel

AU - Tesfalem, Henok

AU - Dorn, Oliver

AU - Fletcher, Adam

AU - Peyton, Anthony

AU - Brown, Matthew

PY - 2022/4/18

Y1 - 2022/4/18

N2 - We report on the use of a novel constrained optimisation algorithm for calibrating the finite element model in an eddy current inspection application. An accurate finite element forward model is often important in such eddy current applications for training neural networks or as part of an iterative solver. However, the subject of calibration of the model has not received much attention in the literature to date.We consider amulti-frequency, eddy current depth profiling application, which is important for non-destructive testing in the nuclear industry. In the optimisation algorithm, we use a Levenberg-Marquardt algorithm and a bisection search to ensure constraints are satisfied, coupled with a truncated Gradient and Hessian method. We calibrate two types of finite element models, one using a filament representation for the coils and the other using a full 3D approach. The results show the feasibility of using the constrained non-linear optimisation algorithm for tuning the finite element model parameters. The mean signal-noise ratio after tuning on a truncated spectrum was 29.47 dB for the 3D model and 28.96 dB for the filament; in contrast the mean SNR using the measured coil parameters was 1.48 dB and 4.98 dB for the two uncalibrated models, respectively. The results show that a filament model is competitive with a 3D model of the coils (within the nuclear graphite application); therefore, a computationally faster filament model can be used with minimal effect on accuracy.

AB - We report on the use of a novel constrained optimisation algorithm for calibrating the finite element model in an eddy current inspection application. An accurate finite element forward model is often important in such eddy current applications for training neural networks or as part of an iterative solver. However, the subject of calibration of the model has not received much attention in the literature to date.We consider amulti-frequency, eddy current depth profiling application, which is important for non-destructive testing in the nuclear industry. In the optimisation algorithm, we use a Levenberg-Marquardt algorithm and a bisection search to ensure constraints are satisfied, coupled with a truncated Gradient and Hessian method. We calibrate two types of finite element models, one using a filament representation for the coils and the other using a full 3D approach. The results show the feasibility of using the constrained non-linear optimisation algorithm for tuning the finite element model parameters. The mean signal-noise ratio after tuning on a truncated spectrum was 29.47 dB for the 3D model and 28.96 dB for the filament; in contrast the mean SNR using the measured coil parameters was 1.48 dB and 4.98 dB for the two uncalibrated models, respectively. The results show that a filament model is competitive with a 3D model of the coils (within the nuclear graphite application); therefore, a computationally faster filament model can be used with minimal effect on accuracy.

U2 - 10.1109/JSEN.2022.3167253

DO - 10.1109/JSEN.2022.3167253

M3 - Article

SN - 1530-437X

JO - IEEE Sensors Journal

JF - IEEE Sensors Journal

ER -

Calibration of a Finite Element Forward Model in Eddy Current Inspection.

Abstract

Access to Document

Fingerprint

Projects

Electromagnetic Sensing Group

Cite this