Ebook: Electromagnetic Non-Destructive Evaluation (XXI)

The ENDE international workshop has been held every year since 1995. Its aim is to bring together engineers and scientists from universities, research institutions and industry active in the research, development and industrial application of Electromagnetic Nondestructive Evaluation.

The 22nd edition of the workshop was held in CEA SACLAY DIGITEO LABS, Saclay, France, September 6–8, 2017. It was co-organized by the CEA LIST Institute and the Laboratoire des Signaux et Systèmes (L2S), with the sponsorship and support of the Digiteo-DigiCosme 2016 program, the University Paris-Saclay, the COFREND organization, and the JSM learned society. It was conducted under the co-chairpersonship of D. Lesselier (L2S) and C. Reboud (CEA LIST), assisted by a Scientific Committee and a Local Organizing Committee, under the auspices of the ENDE International Standing Committee chaired by Professor Zhenmao Chen, School of Aerospace, Xi'an Jiaotong University.

The ENDE topic is crucial to a great number of engineering activities, environmental evaluation and protection issues, and it is recognized for theoretical insight, efficient models and simulations, robust data interpretation, and proper instrumentation combined accordingly. A broad set of themes follows, including non-destructive testing and evaluation (NdT-NdE), static to THz electromagnetics, smart models and high-performance computations, advanced sensors, from design to usage, involving single- or multi-physics, inverse problems, imaging and signal processing in the context of uncertain and relatively scarce data, surrogate models, adaptive databases, model selection, and the qualification of uncertainty, multi-sensor data fusion, towards automatic approaches and decision making, complex material characterization from small to large scale, the monitoring and diagnosis of mechanical structures, and innovative industrial applications.

Ten oral sessions were organized at ENDE 2017during the two-and-a-half days of the workshop: high frequency applications I & II; instrumentation I and II; material characterization I & II; theory and modeling; eddy current testing; low frequency inversion; and time domain methods. Two poster sessions enabled very interesting discussions in a more informal fashion. A total of 114 participants were registered (from Europe, the United States and Canada, China, Korea, Japan, and Thailand), and 3 invited keynotes, 47 oral contributions and 43 poster contributions were given in all, short versions of all such contributions being published in the Workshop Digest.

The first keynote was entitled “Product Uniformity Control – A Research Collaboration of European Steel Industries to Non-Destructive Evaluation of Microstructure and Mechanical Properties” and was given by Frenk van den Berg, Principal Scientist at Tata Steel, R&D IJmuiden, Netherlands. The second keynote, “Intrinsic modeling and inversion of ultra wideband microwave radar data for nondestructive testing of soils and materials”, was given by Sébastien Lambot, Professor in Agricultural and Environmental Engineering at UCL & FNRS Scientist, Louvain-la-Neuve, Belgium, and the third, “Numerical tools for the ultrasonic Non Destructive Testing of anisotropic plates”, was given by Sonia Fliss, Assistant Professor at ENSTA ParisTech, member of the Poems Laboratory (UMR 7231 CNRS-INRIA-ENSTA), Palaiseau, France.

On condition of being first authors and attending the workshop, students were proposed for the best presentation (oral or poster). The first prize went to Cesar Camerini for the communication he co-authored with João Marcos Rebello, Rafael Wagner Santos, João Marcio Santos, Tomasz Chady and Gabriela Pereira, entitled “Eddy Current System for Clad Pipe Inspection”. The second went to Takanori Matsumoto for the communication he co-authored with Tetsuya Uchimoto, Toshiyuki Takagi, Gabor Vertesy, Hidehiko Kage and Gerd Dobmann, entitled “Nondestructive Evaluation of Mechanical Properties of Cast Iron by Magnetic Incremental Permeability Method”, and the third prize went to Almpion Ratsakou for the communication he co-authored with Christophe Reboud, Anastassios Skarlatos and Dominique Lesselier, entitled “Fast Models Dedicated to the Simulation Of Eddy Current Thermography”.

In addition to an array of premier contributions, the 21 previous ENDE have been marked by edited peer-reviewed proceedings published by IOS Press as Electromagnetic Nondestructive Evaluation, within the series Studies in Applied Electromagnetics and Mechanics. The first (with the slightly different title of Non-Destructive Testing of Materials) appeared in 1995 under R. Collins, W.D. Dover, J.R. Bowler, and K. Miya. Twenty more followed. This suite of proceedings provides a unique perspective on the many challenges and advances of electromagnetic non-destructive evaluation, and adds significantly to the expert exchanges that have taken place at each workshop. The 22nd proceedings, following ENDE 2017, are proposed herein, as an Open Access publication on the IOS Press e-book platform, including 1 keynote and 34 peer-reviewed extended contributions.

To conclude, the organizers would like to thank all those Committees members who have tirelessly contributed to the workshop, and the many reviewers of the extended papers who have played a key role in the making of these proceedings.

Saclay, March, 2018

D. Lesselier, L2S

C. Reboud, CEA LIST

Editors

A steel plate having hardened and non-hardened regions in one plate is promising material in automobile field owing to its high performance. This steel plate has to be evaluated whether the non-hardened region was prepared properly at the desired position with desired hardness. However, the conventional evaluation method of hardness measurement is destructive testing and requires a long inspection time. Therefore, a reliable and fast nondestructive inspection method is required. In this study, the hardened and non-hardened regions were evaluated using an eddy current testing (ECT) to distinguish each region. The measured resistivity and permeability at each region were different and these differences were found to be caused by the crystal structure change of heat treatment. Therefore, the magnetic response between the hardened and non-hardened regions was different and this difference was correlated with the hardness of each region. These results indicate that the hardened and non-hardened regions can be distinguished using ECT.

In the process of magnetic flux leakage (MFL) testing, the leakage magnetic field decreases as the testing velocity increases, which leads to the severe detection omission. To address this issue, factors affecting high-speed magnetic flux leakage testing are analyzed, multistage magnetization structure is then proposed. By comparing magnetic flux leakage signals characteristics on the surface of steel pipe with high speed and low speed, effectiveness and feasibility of multistage magnetization structure are analyzed. Results show that multistage magnetization structure can effectively extend magnetization time. Compared with single magnetization, the magnetic flux leakage signal of the inner wall of the steel pipe can be detected effectively with multistage magnetization structure at high speed, and the magnetic flux leakage testing precision is improved.

Velocity induced eddy current inspection is a technique to inspect conductive materials. It has a good detection performance and is particularly well suited to situations where the speed is a factor to be considered because its sensibility increases with the moving speed. This paper assesses the field produced by several probe geometries in the detection of linear machined defects. A more detailed analysis is carried out for a probe that includes an array of cubic magnets to induce the eddy currents inside the sample and Hall sensors to measure the resultant magnetic field.

This paper presents an efficient strategy to simulate pulsed eddy current testing signals. It relies on adaptive sparse grids interpolation for the fast evaluation of Fourier harmonics composing the transient pulsed eddy current signal. In addition to the discussion on the performance obtained, some useful post-processing procedures enabling to retrieve relevant information from raw signals are described.

We present a level set technique for 3D Magnetic Induction Tomography with an emphasis on applications to the screening of small boxes up to cargo containers. A level set method is used for modeling a shape evolution when minimizing a given cost functional. Numerical results will be presented that illustrate the performance of our method in practical situations. A novel line-search technique is introduced that is suitable to control the shape evolution for this computationally demanding MIT inverse problem.

Austenitic stainless biomaterials are worldwide used in biomedical practice. Electromagnetic nondestructive evaluation allows their contactless investigation in terms of magnetic field mapping. Intrinsic magnetic field directly reflects changes in their mechanical properties. This article presents investigation of concrete austenitic biomaterials by use of fluxgate sensor. Pre-defined plastic deformation levels of the cylindrical specimens are evaluated. Three-dimensional scanning procedure has been performed.

In order to obtain accurate quality control of steel products, it is desirable to be able to monitor the mechanical properties non-destructively. This study proposes using an electromagnetic (EM) sensor, suitable for use on strip samples, as a tool for non-destructive steel characterisation, in particular phase balance and grain size in dual phase (DP) steels and hence strength. It is known that the low frequency inductance measured using an EM sensor depends on the relative permeability of the sample and that the permeability is affected by microstructural features (i.e. phase fraction/distribution and, to a lesser extent, grain size are the important features in DP steel). EM sensors can be used to characterise austenite and ferrite fraction in hot strip mills (EMspec^TM system) and for statistical correlations to mechanical properties (IMPOC and HACOM systems) in cold strip mills. In this paper an EM sensor system and a finite element based sensor – microstructure model have been used to characterize DP microstructures (using commercial DP600, DP800 and DP1000 grades and a heat treated DP600 grade) taking into account the effect of strip thickness on the signals. This paper also reports on the relationship between the ferrite fraction and tensile strength, which follows the expected relationship from the literature, and the magnetic permeability (determined from the EM sensor signal which is directly related to ferrite fraction and grain size) and tensile strength. The ability of the EM sensor to determine the tensile strength is therefore illustrated taking into account strip thickness and microstructure.

We design a linear sampling method to retrieve the shape of an anomaly in microwave imaging. It is is based on the structure of nonzero singular vectors associated with the nonzero singular values of a complex-symmetric matrix, whose elements are measured S-parameters. Simulation results shows the effectiveness and limitations of the linear sampling method in real-world microwave imaging.

Chemiluminescence anlaysis and high-frequency eddy current testing (HF-ECT) were examined as nondestructive evaluation techniques to assess the hardening degree of epoxy resin in carbon fiber-reinforced plastic. Both techniques showed a correlation with the hardening degree of epoxy resin measured by differential scanning calorimetry. The results also indicated that the application of HF-ECT at frequencies of >50 MHz is suited for hardening degree determination.

Orthogonal eddy current sensors operating in differential mode was applied to evaluate fatigue cracks in clad pipelines circumferential welds. A dedicated electronic hardware was developed to drive the sensors and measure the electrical impedance. In the preliminary experiments, an automated inspection was performed with the goal to evaluate sensors detectability and different scanning speed was tested to reproduce in service situation. The results have shown that the presented eddy current transducer is a potential solution for a fatigue crack detection on a clad circumferential weld bead.

This paper presents advanced methods using the deep X-ray lithography (X-ray LIGA) and the powder metallurgy processing to fabricate imitative Stress Corrosion Crack (SCC) for electromagnetic NDE as needed to be a reference specimen. A pattern of the imitative SCC was formed by using X-ray LIGA with controllable shape and size with precise details of SU-8 material. The prepared pattern was then inserted into mixed-stainless steel and Sn powders in a soft mold by using less pressure filling method. Afterward, the specimens were sintered at 1100°C. The results showed that these promising combined techniques can be utilized to fabricate micro sizes of the imitative SCCs. In addition, an inversion scheme was utilized to evaluate the equivalent conductivity of the specimens. The analysis results revealed that they were appropriated for desired equivalent conductivities of actual SCC detection by an eddy current testing.

This manuscript addresses the “classical” problem of estimating electrical conductivity of metals. The novel contribution of this work is the use of a new set of features, namely time constants of the source-free response in a pulsed eddy current testing (PECT) experiment. Time constants characterize the source-free response and increase monotonically with the electrical conductivity of the specimen. Time constants form a particularly important set of features because they do not depend upon the probing system and, hence, they are not sensitive to probe lift-off and tilting which are responsible for significant experimental errors.

A sparse grid surrogate model using hierarchical B-spline basis functions is used to approximate the objective function in an optimization-based inversion algorithm. The B-spline basis provides a smooth interpolant of the objective function and the gradient of the interpolant is readily available in closed-form. The latter is used in a gradient-based minimum search algorithm that results in the approximate solution of the inverse problem. The method is computationally more efficient than using gradient-free direct search methods, as illustrated by an example drawn from eddy-current nondestructive testing.

Due to corrosive and hostile environment, in-service nonmagnetic pipes are prone to such anomalies as the external corrosion which poses a severe threat to structural integrity. Gradient-field Pulsed Eddy Current technique (GPEC) has been found superior to other non-destructive evaluation (NDE) techniques particularly pulsed eddy current testing in evaluation and imaging of subsurface corrosion in planar conductors. In this paper, GPEC for evaluation and imaging of external corrosion in nonmagnetic pipes is investigated. Two orthogonal gradient-field signals are simultaneously acquired and used for visualization of external corrosion. Through experimental investigation, it has been found that: (1) the depth of the external corrosion can be implied by the acquired corrosion images; and (2) the corrosion boundaries can be detected and identified via GPEC.

This study is concerned with void detection of steel products using electromagnetic acoustic transducer (EMAT). The dynamical behaviors of the pulser-receiver EMAT measurement system are described by a transient eddy current analysis. The ultrasonic wave propagation inside steel products is then given by an initial-boundary elastic problem defined on the steel domain with unknown void parameters. Thus the parameter-to-output mapping is developed by time dependent finite element model with unknown parameters. An inversion methodology is proposed using computational intelligent approach based on the reduced order expansion. Results of computational experiments to demonstrate the efficacy of the proposed method are reported.

In steel manufacturing, the conventional method to determine the mechanical properties and microstructure is by offline, destructive (lab-)characterisation of sample material that is typically taken from the head or the tail of the coil. Since coils can be up to 7 km long, the samples are not always representative for the main coil body. Also, the time delay (typically a few days) between the actual production and the availability of the characterisation results implies that these results cannot be exploited for real-time adaptation of the process settings.

Information about the microstructure and material properties can also be obtained from electromagnetic (EM) and ultrasonic (US) parameters, which can be measured in real-time, non-destructively, and over the full length of the steel strip product. With the aim to improve the consistency in product quality by use of inline EM and US measurements, a European project called “Product Uniformity Control” (PUC) has been set up as a broad collaboration between 4 major European Steel Manufacturers and 10 Universities/Research institutes.

Using both numerical simulations and experimental characterisations, we study the inline measured EM and US parameters in regard of the microstructural and mechanical properties. In this way, we aim to establish an improved understanding of their mutual relationships, and to apply this knowledge in existing and new non-destructive evaluation techniques.

In this paper, the concerted approach of modelling and experimental validation will be addressed, and results of this work will be shown in combination with inline measured data.

In recent years, aging and deterioration of social infrastructure has become a major problem. A non-destructive testing technique for early detection of degradation (such as internal corrosion) is required to extend the lifetime of these structures. Currently, there are various methods used for inspection. Among these, non-destructive testing using magnetic fields is widely used due to advantages such as ease of use and high speed. However, magnetic inspection is generally employed only for inspection of the surface and subsurface, since it is difficult to inspect internal corrosion using this technique. Therefore, we have developed a device for detecting thinning of steel plates due to corrosion using multiple extremely-low frequencies and a high-sensitivity magnetic sensor. However, it is necessary to sweep multiple frequencies to obtain a frequency spectrum, and this is very time-consuming, especially in extremely low-frequency ranges. In this study, we have developed an analysis method based on a multiple-frequency applied magnetic field and FFT (fast Fourier transform) analysis of the detection signal. Thus, we have successfully reduced the measurement time while maintaining high detection accuracy.

In this communication, we present a design methodology for an insensitive radiating structure for concrete health monitoring applications. It takes advantage of an optimized smaller vacuum box surrounding a patch antenna. The Ansys HFSS software is used to simulate the studies of different configurations. The patch antenna consists of a radiating metallic element with a ground plane printed on a low-cost FR4 substrate. It is optimized to be linearly polarized and to be able to operate in free space medium at the first two resonant frequencies of 1.55 GHz and 1.95 GHz. This study demonstrates that a 5 mm high vacuum box above main radiating antenna, is sufficient to reduce the influence of three surrounding dielectric materials studied in this work, such as a PVC, concrete and limestone. Theoretical predictions, for all types of surrounded materials, show reasonably good agreement with experimental results.

As a kind of super-light lattice material, the sandwich plate with pyramidal truss cores has wide application prospect. In this paper, a vibration parameter based nondestructive testing method using Uniform Load Surface (ULS) curvature was proposed to detect the delamination defects of the pyramid-type lattice sandwich plate. The feasibility of the approach is that the surface smoothing method and curvature mode change rate method are adopted to overcome the dependence on the unflawed model. The validity and efficiency of the proposed methods were demonstrated through numerical simulations.

The global sensitivity analysis of electromagnetic nondestructive evaluation (NDE) by means of Sobol' indices are considered in this work. To reduce the computational burden, a sparse grid surrogate model is used. The latter can simply replace the true simulator to some extent, but it can also be used to numerically evaluate the integrals defining the Sobol' indices. In most of the NDE setups, the output is not a scalar quantity but functional data (e.g., a surface scan); a method is presented to take this into account. The sparse grid based sensitivity analysis is compared to classical techniques via examples drawn from electromagnetic NDE.

This paper reports an experimental study to detect and distinguish surface and sub-surface cracks in non-ferromagnetic materials using eddy current testing with a multi-harmonic excitation signal. A minimization of the crest factor of the excitation signal was done using a Shapiro-Rudin phase sequence. The chosen range of frequencies was based on a corresponding range of standard depths of penetration (SDP) in order to ensure that the eddy currents penetrate more or less an aluminum plate with 4 mm of total thickness. Experimental tests were realized on surface and sub-surface cracks with depths of 0.5, 1, 1.5 and 2 mm. The results obtained by the ECT probe are analyzed in the frequency domain.

The modelling of the eddy current testing of U-bend steam generator tubes is a valuable tool for interpreting measured signals due to geometric distortions that can arise during the bending process. Especially, the decentred trajectory of the probe into the bending must be taken into account to ensure the accuracy of the simulated signal. As the exact trajectory is a priori not known, we propose an efficient way to build a database of the simulated signal for all possible positions of the probe into the tube, based on an adaptation of the boundary element method that is developed at CEA LIST.

This communication presents the first development aiming at efficiently simulating configurations of eddy current thermography for nondestructive evaluation. The numerical method proposed here is based on the Finite Integration Technique for both electromagnetic and thermal problems. Simulation results obtained using two different materials, steel and aluminum, are compared and discussed with respect to the presence of a flaw affecting the piece under test.

Metal Magnetic Memory (MMM) technique has been extensively used as a qualitative method to test the position of possible damages in ferromagnetic metal structures, a bottleneck in further development of MMM technique is the quantitative relationship between MMM signal and damage characteristic has not been fully revealed. In this paper, analysis of influences of stress and discontinuous structure on MMM signals is presented. Samples of diverse types of structural defects were made and then loaded with different series of stress. The resulting residual deformations and gradients of the magnetic field were measured. Analysis of the quantitative relationship between MMM signals and stress was conducted. The results show: 1) The magnetic field gradient above the structure-continuous zone increases with stress. Therein lies the potential to estimate the stress level point by point. 2) The second-order gradient of the magnetic field near the structure-discontinuous zone increases with the maximum strain in a similar way for different types of discontinuous structures, implying a possibility that the second-order gradient of the magnetic field may be conceive some information about the maximum stress at the edge of discontinuous structures independent of their geometry characteristics.