Kriging for Eddy-Current Testing Problems

Bilicz, Sándor; Vázquez, Emmanuel; Gyimóthy, Szabolcs; Pávó, József; Lambert, Marc

doi:10.1109/tmag.2010.2043418

Cited by 13 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The database combined with a simple interpolator (here a nearest neighbour) can then be used as an improved fast forward surrogate model. Though the preliminary results are encouraging, such a basic interpolator is however not yet optimal and some other techniques should be used [27].…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Kriging-based generation of optimal databases as forward and inverse surrogate models

2010

View full text Add to dashboard Cite

Numerical methods are used to simulate the mathematical models of a wide range of engineering problems. The precision provided by such simulators is usually fine but at the price of computational cost. In some applications this cost might be crucial. This leads us to consider cheap surrogate models in order to reduce the computation time still meeting the precision requirements. Among all available surrogate models, we deal herein with the generation of an "optimal" database of pre-calculated results combined with a simple interpolator. A database generation approach is investigated which is intending to achieve an optimal sampling. Such databases can be used for the approximate solution of both forward and inverse problems. Their structure carries some meta-information about the involved physical problem. In the case of the inverse problem, an approach for predicting the uncertainty of the solution (due to the applied surrogate model and/or the uncertainty of the measured data) is presented. All methods are based on kriging-a stochastic tool for function approximation. Illustrative examples are drawn from eddy-current nondestructive evaluation.

show abstract

Section: Resultsmentioning

confidence: 99%

“…The entries of Q are easy-to-compute, as Q i (x j ) = y j (t) − y i (t) y . By using Q, the prediction (27)…”

Section: A Summary Of Ordinary Krigingmentioning

confidence: 99%

Kriging-based generation of optimal databases as forward and inverse surrogate models

2010

View full text Add to dashboard Cite

show abstract

“…A kriging modeling technique with multiple design variables and a strong nonlinearity is used to set the objective function. Also, the kriging modeling technique can accurately interpolate sample data and it is also possible to model multiple local poles [26,27].…”

Section: Optimal Design Of the Fcpm Modelmentioning

confidence: 99%

Optimal Design of a Novel Consequent-Pole Interior Permanent Magnet Motor with Flared-Structured Rotor

Yoon,

You

2024

Applied Sciences

View full text Add to dashboard Cite

Interior permanent magnet motors are widely used in applications requiring high power density and high efficiency due to their high torque-generating capabilities. Recently, given the price fluctuations and unstable supply of rare earth permanent magnets, alternative configurations with reduced use of permanent magnets are being sought. Among the various candidates related to this, the consequent-pole type rotor structure can halve the number of permanent magnets used compared with conventional structures. However, in a no-load analysis, the waveform of the back electromotive force becomes asymmetric, generating a harmonic component. As a result, there is a disadvantage that the torque ripple increases. To overcome these shortcomings, we propose a novel rotor structure that applies a consequent-pole structure to an embedded permanent-magnet motor structure, wherein a number of permanent magnets are arranged in a flared structure to constitute a single polarity. In the proposed flared-structured magnet arrangement, it is possible to adjust the angle of the permanent magnet and the polar angle to mitigate the asymmetry of the back-EMF waveform. The proposed structure was optimized with a genetic algorithm and a prototype of the optimal model was constructed and experimentally evaluated to verify its validity. Finally, the performance improvement and validity of the proposed structure were verified by comparing the analysis results of the optimal model with the experimental results.

show abstract

“…As for the interpolation, multi-linear [5], kriging [8] or sparse-grid [6] methods have been proposed in our previous work. Multi-linear interpolation is fast in terms of computation for a small number of unknowns, kriging is more accurate but requires the pre-estimation of the covariance matrix while sparse-grid is more suitable to high dimensional problems.…”

Section: Data-fitting Metamodelmentioning

confidence: 99%

Metamodel-based Markov-Chain-Monte-Carlo parameter inversion applied in eddy current flaw characterization

Cai

Miorelli

Lambert

et al. 2018

NDT & E International

Self Cite

View full text Add to dashboard Cite

Flaw characterization in eddy current testing usually requires to solve a non-linear inverse problem. Due to high computational cost, Markov Chain Monte Carlo (MCMC) methods are hardly employed since often needing many forward evaluations. However, they have good potential in dealing with complicated forward models and they do not reduce to only providing the parameters sought. Here, we introduce a computationally-cheap surrogate forward model into a MCMC algorithm for eddy current flaw characterization. Due to the use of a database trained off-line, we benefit from the MCMC algorithm for getting more information and we do not suffer from the computational burden. Numerous experiments are carried out to validate the approach. The results include not only the estimated parameters, but also standard deviations, marginal densities and correlation coefficients between two parameters of interest.

show abstract

Kriging for Eddy-Current Testing Problems

Cited by 13 publications

References 12 publications

Kriging-based generation of optimal databases as forward and inverse surrogate models

Kriging-based generation of optimal databases as forward and inverse surrogate models

Optimal Design of a Novel Consequent-Pole Interior Permanent Magnet Motor with Flared-Structured Rotor

Metamodel-based Markov-Chain-Monte-Carlo parameter inversion applied in eddy current flaw characterization

Contact Info

Product

Resources

About