Pecreview PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 16

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/318361772

Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review

Article  in  Chinese Journal of Mechanical Engineering · April 2017


DOI: 10.1007/s10033-017-0122-4

CITATIONS READS

43 1,837

3 authors, including:

Ali Sophian Mengbao Fan


International Islamic University Malaysia China University of Mining and Technology
42 PUBLICATIONS   1,355 CITATIONS    75 PUBLICATIONS   532 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Novel Sensing Network for Intelligent Monitoring (NEWTON) View project

Adaptive Sparsity Micro-crack Separation and Quantification Using Electromagnetic Thermography View project

All content following this page was uploaded by Ali Sophian on 17 October 2017.

The user has requested enhancement of the downloaded file.


Chin. J. Mech. Eng. (2017) 30:500–514
DOI 10.1007/s10033-017-0122-4

REVIEW ARTICLE

Pulsed Eddy Current Non-destructive Testing and Evaluation:


A Review
Ali Sophian1 • Guiyun Tian2,3 • Mengbao Fan4

Received: 2 November 2016 / Revised: 18 January 2017 / Accepted: 28 March 2017 / Published online: 17 April 2017
Ó Chinese Mechanical Engineering Society and Springer-Verlag Berlin Heidelberg 2017

Abstract Pulsed eddy current (PEC) non-destructive test- 1 Introduction


ing and evaluation (NDT&E) has been around for some
time and it is still attracting extensive attention from Despite its approximately-five-decade-long history, PEC is
researchers around the globe, which can be witnessed still considered by many as a new emerging eddy current
through the reports reviewed in this paper. Thanks to its NDT&E technique. Compared to other eddy current testing
richness of spectral components, various applications of (ECT) techniques this view can be true. Literature shows
this technique have been proposed and reported in the lit- that PEC has been attracting the attention of researchers
erature covering both structural integrity inspection and from around the globe, including countries, such as China,
material characterization in various industrial sectors. To UK, Canada, Portugal, USA, South Korea, Japan, France,
support its development and for better understanding of the Slovakia, Poland, and Italy.
phenomena around the transient induced eddy currents, The amount of attention that PEC NDT&E has been
attempts for its modelling both analytically and numeri- receiving owes to the key potential benefits that it offers.
cally have been made by researchers around the world. The first and main advantage is that, compared to single
This review is an attempt to capture the state-of-the-art frequency ECT, PEC inherently has a broadband of fre-
development and applications of PEC, especially in the last quencies [1], which is advantageous for any eddy-current-
15 years and it is not intended to be exhaustive. Future based NDT&E techniques due to the frequency-dependant
challenges and opportunities for PEC NDT&E are also skin effect. Another benefit is that PEC signals are rela-
presented. tively easier to interpret, while it requires a special skill of
the operators for interpreting conventional ECT signals
Keywords Non-destructive testing  Pulsed eddy currents  which are presented in the impedance plane trajectory.
Material characterization  Structural integrity  Conventional ECT only applies a single frequency for
Non-destructive evaluation excitation which makes it unable to detect both surface and
sub-surface defects reliably. The improved technique is the
multi-frequency ECT which applies different excitation
frequencies, one after another. Compared to multi-fre-
& Guiyun Tian quency ECT, PEC can potentially be applied in shorter
[email protected]
time for inspection of different depths as PEC applies a
1
Faculty of Engineering, International Islamic University wideband of frequencies in a single pulse. This allows to
Malaysia, Kuala Lumpur, Malaysia reduce the measurement time to the minimum one
2
School of Automation Engineering, University of Electronic depending on the sample characteristics. Fig. 1 provides
Science and Technology of China, Chengdu, China the illustration of the excitation waveforms of each of the
3
School of Electrical and Electronic Engineering, Newcastle methods.
University, Newcastle upon Tyne NE1 7RU, UK Similar to other ECT techniques, in general PEC
4
School of Mechatronic Engineering, China University of requires no surface preparation which leads to reduction of
Mining and Technology, Xuzhou 221116, Jiangsu, China inspection time and costs efficiency is improved. The

123
Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review 501

Conventional ECT
1

-1
0 5 10 15 20
Multi-frequency ECT
1 1 1

0 0 0

-1 -1 -1
0 10 20 0 10 20 0 10 20
PEC
Fig. 2 Illustration of the working principle of ECT
1
rffiffiffiffiffiffiffiffi
2
0.5 d¼ ; ð1Þ
-lr

0 where d is skin depth (m), l is magnetic permeability (H/


0 5 10 15 20
Time m), r is electrical conductivity (S/m) and x is angular
frequency (rad/s). The equation shows that the depth of
Fig. 1 Illustration of excitation waveforms for different ECT penetration depends on the excitation frequency. The lower
techniques
the frequency, the deeper the penetration and vice versa. In
contrast to conventional sinusoidal eddy current technique,
inspection can also be done without interrupting the oper- where the excitation is limited to one frequency compo-
ation or service of the structure being tested, unlike for nent, pulsed eddy current techniques excite the induction
example X-ray testing. In many applications where the coil with a pulse waveform. The frequency components of
sample is coated, no removal of the coating is required pulse waveform can be demonstrated using Fourier
when ECT NDT&E is used. Any eddy-current systems are Transform. If the excitation waveform is defined as
relatively cost-effective and reliable. 8
> T T
In the following sections, the concept of PEC is briefly < A;   t  ;
f ðtÞ ¼ 2 2 ð2Þ
discussed which is then followed by the review in systems,
: 0; jtj [ T ;
>
modelling, signal processing and applications. A conclu- 2
sion completes this review paper.
where A is the amplitude of the pulse and T is the pulse
width, then using the amplitude spectrum of the excitation
is defined as
2 Concept of Pulsed Eddy Current
2 sin xT=2
F ðx Þ ¼ : ð3Þ
In eddy current NDT, an AC-driven excitation coil induces x
eddy current in the sample through electromagnetic cou- Fig. 3 shows examples of the pulses with two different
pling. In turn, the circulation of the eddy current induces a widths and their power spectra, which shows that the
secondary magnetic field as illustrated Fig. 2. This field excitation has a series of frequency components, which has
will vary if flaw that impedes the eddy currents is present given the technique the potential to inspect different depths
or there is a change in the electrical conductivity, magnetic simultaneously and therefore it will be able to offer more
permeability or thickness of the sample. The change in the information compared to the conventional approach.
field will be picked up by a sensing device, which is typ-
ically either a coil or a magnetic sensor.
The penetration and the density of the eddy current in 3 PEC Systems
the sample is an important issue in any ECT. The pene-
tration is limited due to the skin effect, which causes its Despite variations that exist, a typical PEC system will
density to decrease exponentially with depth. The depth at look like the illustration shown in Fig. 4. A pulse signal at
which the density has reduced to 1/e of the density at the a chosen frequency and pulse width is generated which is
surface is termed the skin depth d and defined by then power-amplified to drive an excitation coil. In turn, a

123
502 Ali Sophian et al.

(a) 1 Pulse width = 2 ms (b) 1

Pulse width = 5 ms
Pulse width = 2 ms
0.8 0.8
Amplitude (V) Pulse width = 5 ms

Amplitude
0.6 0.6

0.4 0.4

0.2 0.2

0 0
-20 -15 -10 -5 0 5 10 15 20 0 50 100 150 200
Time (ms) Frequency (Hz)

Fig. 3 (a) Examples of pulses with different widths, (b) Power spectra of the pulses

Fig. 4 Generic configuration of


a PEC NDT system

time-varying magnetic field is induced by the current in the duration in order not to overheat the coil and the driver
excitation coil. The magnetic field, which is called the pri- electronics. And there are also other shapes of excitation
mary field, induces eddy current in the sample. Consecu- signal that have also been used and proposed by
tively, a secondary magnetic field is induced by the eddy researchers. A study on different excitation waveforms,
current and it opposes the primary field. This secondary field namely square, half-sine and ramp, shows a favour for
is then detected by a sensing device, which typically can be the square waveform [2]. A variable pulse width exci-
either a magnetic sensor or a coil. The output signal of the tation has also been proposed [3], which was used in the
sensing device is then passed to the next stage to be con- inspection of subsurface corrosion in conductive struc-
ditioned and processed where eventually features are tures [4]. Pulse width modulation, as illustrated in Fig. 5,
extracted in order to infer the desired parameters, such as provides different frequency spectra and is suggested of
wall thickness and lift-off, from the testing. being able to eliminate the need for reference sample
From one implementation to another, the systems vary signal [5]. PEC has also been implemented by using the
primarily because of the differences in the excitation sig- decaying part of the step signal, rather than the raising
nal, excitation system, sensing device and the signal pro- part, after the power supplied to the excitation coil is
cessing and feature extraction techniques. These variations disconnected [6–8].
are discussed below.
3.2 Probes
3.1 Excitation Signals
Typically, a PEC probe would contain one excitation coil
In many implementations, the excitation current or and one or more sensing devices. An excitation coil gen-
voltage is a square waveform. In some other applica- erates primary transient excitation field, while one or more
tions, the excitation is of rectangular waveform which sensing devices picks up secondary eddy current field due
allows a very high power to be delivered in a limited to a sample. Probe designs are usually optimized in terms

123
Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review 503

and therefore no reference signals are required. This type


of probe can be implemented by using two pick-up coils or
two magnetic sensors with the output signal being the
difference between the two output signals from the sensing
devices. A differential double-D probe using two Hall
devices has been investigated by Park, et al [13] which
shows a potential for detection and sizing of sub-surface
cracks in stainless-steel structures. Differential probes have
also been studied for crack detection near a fastener in
aircraft structures [14], [15].
Less common probe configurations have also been used,
such as a planar matrix probe that can generate a color map
that is useful in the identification of defects [16]. Their
work shows the use of an 8-by-8 array of sensors, as shown
in Fig. 7(a), successfully maps the surface defects that have
been artificially made on the sample, which justifies the
Fig. 5 Excitation currents with varied pulsed widths [5]
complexity of the excitation and sensing circuits used in
of its structure, the type of sensing elements and the use of the probe. The application of independent excitations lead
cores based on the specific applications in which they will to a more uniform excitation field which, in turn, leads to a
be deployed. simpler interpretation of the detected signals. Another
Based on the electromagnetic coupling between the interesting example of the use of sensor array in ECT, is
excitation coil and the sample, eddy current probe’s excita- shown in Fig. 7(b), where the printed array is flexible and
tion coils can generally be categorized into one of the fol- can be used to produce a color map of surface corrosion
lowing three types: surface (or pancake) coil, encircling coil [16]. Another unique example is a symmetric excitation
(or OD for outer diameter) and internal coil (also called coil introduced by Yang et al, which is expected to gen-
bobbin or ID for inner diameter) [9]. The three types of coils erate linear eddy currents with the benefit of virtually no
are illustrated in Fig. 6. Surface or pancake coils may be field will be detected by the pick-up sensor when no defect
orientated either parallel or normal to the surface of the is present [18].
sample and they are used for both flat and curved samples. Another differentiating feature is the shape of the coil.
Encircling coils are generally used in the inspection of Rather than being circular, which is the most common
cylindrical elongated structures, such as hollow pipes and shape, the coil may be rectangular or racetrack. This non-
solid rods [10], [11]. Coils of this type form a circle around circular type of a coil is also referred as directional as
the diameter of the test object coaxially. The specimen opposed to non-directional or isotropic for circular coils.
maybe insulated or coated. The bobbin-typed coils are usu- With directional probes, the paths of the induced eddy
ally used to inspect hollow cylindrical structures, such as currents are not circular, and, therefore, they are more
pipes and bore holes, from the inside. In PEC NDT, a coil of sensitive to changes in a particular direction. One example
this type which is used in remote field mode has been used for of the use of directional rectangular coil is the work done
measurement of wall thickness of ferromagnetic tubes [12]. by He, et al [19], [20]. They also state that a more uniform
Differential probes, as opposed to absolute probes, are eddy current distribution is being an advantage gained by
also used with the advantage of the self-nulling features using such a coil.

Fig. 6 Coil types used in ECT: (a) surface coil, (b) encircling coil, and (c) internal coil

123
504 Ali Sophian et al.

3.3 Sensing Devices

While the excitation is always achieved by using an


induction coil, the sensing devices can broadly be cate-
gorised into two different types, namely induction coils and
magnetic sensors. The outputs of induction coils and
magnetic sensors are given by
dB
Vcoil ¼ N  A  ; ð4Þ
dt
Vmag sen ¼ K  B; ð5Þ
where N is the number of the coil turns, A is the area that
magnetic flux passes through, B is the magnetic field
density, and K is a coefficient of the magnetic sensor.
The output signal of an induction coil depends on the
rate of change of magnetic flux density, while that of a
magnetic sensor is directly proportional to magnetic flux
density. As a result, the response from an induction coil
exhibits similar characteristics with that of a magnetic
sensor [23]. Each of these types has its own advantages and
disadvantages. Induction coils play a dominant role in use
for simplicity of operation and design, wide frequency
bandwidth and large dynamics. However, induction coils
are only sensitive to AC magnetic field, which is one of the
drawbacks, although this drawback could be handled by
introducing movement to the coil [24]. Magnetic sensors
are typically sensitive to low frequency signals and offer a
better spatial resolution. Hence, magnetic sensors have
gained wide acceptance for enhancement of sensitivity and
spatial resolution in sensing low frequency magnetic field
[25]. In the applications where extremely high tempera-
tures are involved, induction coils can be the only feasible
option.
Fig. 7 Examples of arrayed sensors: (a) A planar matrix probe [16], There are various types of magnetic field sensors avail-
(b) A flexible printed circuit array produced by Southwest Research
Institute (adopted from [17]) able nowadays. Accumulatively, they can sense field from
as low as in the order of several 10-15 T up to around 100
mT [26]. In PEC NDT, the most widely used magnetic
A ferromagnetic core and shielding maybe present with sensors are Hall effect devices (e.g., Ref. [27]) and mag-
the purpose of concentrating the magnetic flux and netoresistive (MR) devices. Both of these sensor types are
amplifying the signal. Hence, for enhancement of the relatively low cost. Hall effect devices have the highest
sensitivity of the probe in certain applications, a ferro- dynamic range, covering approximately 100 lT up to 100
magnetic core or shielding is added to the probe. An mT, however they have high noise, low resolution and low
example of the study in this topic is introduced in Ref. [21], sensitivity.
where the sensitivity in sub-surface detection was evalu- Out of three existing types of MR technologies, two
ated. The study, which uses both numerical modelling and types are used in PEC NDT, which are anisotropic mag-
experimental tests, shows a favour for a ferrite-cored but netoresistance or AMR(e.g., Ref. [28]) and giant magne-
unshielded over a shielded, ferrite-cored probe. A study by toresistance or GMR (e.g., Ref. [29]). The use of the latest
Zhou, et al [22] shows that a shield made of iron, that is development of MR, i.e., tunnelling magnetoresistance
positioned between the excitation coil and the magnetic (TMR), in PEC systems has not been found by the authors
sensor, provides a higher sensitivity compared with non- in any literature despite it has the highest sensitivity. The
shielded and aluminium-shielded probes for ferromagnetic lack of the presence if TMR in PEC systems may be due to
inspection. its relatively new availability. While its technology is still

123
Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review 505

being improved, an analysis – that may be rather outdated models for PEC since 1980’s [36–39], with the initial
now - shows that despite its higher sensitivity compared objective of gaining better understanding of the PEC phe-
with AMR and GMR, its detectivity is not necessarily nomena when used in NDT inspection. Therefore, the
higher due to its higher 1/f noise [30]. The detectivity relative success in the modelling of PEC has allowed the
signifies the lowest magnetic field a sensor can detect, advancement of PEC and its successful application to many
where the signal-to-noise ratio is unity. However, it is different problems. In the following paragraphs, a few
expected that further development of TMR will improve its examples of works dealing with PEC modelling are men-
detectivity and may make it more interesting for PEC tioned with no intention for delving into details.
applications. Recently, Desjardins, et al have presented an analytical
Another type of magnetic field sensor, namely high approach [10], [40] to model an encircling coil used around
temperature superconducting quantum interference device a ferromagnetic rod. The method computes the time-do-
(HTS SQUID) magnetometer has also been studied in PEC main electromagnetic response of the system and the
[31]. SQUID has the highest sensitivity with the ability of results were validated by experimental data that display a
measuring in the range down to 10-15 T, however its very good agreement. Increasingly, a method for deter-
applications in NDT is still limited due to its practicality mining the electrical conductivity and the magnetic per-
and costs. meability that arises from the analytical solution for the
Whilst most of sensing devices used are absolute, the ferromagnetic rod and encircling coils has been also sug-
use of gradient-field GMR sensors has also been reported gested by Desjardins et al [11]. This is a good example of
by Li, et al [32]. Interestingly, they discover that the gra- how modelling may enhance the use of PEC in NDT
dient-field measurements give better sensitivity and accu- inspection.
racy in detection and sizing artificial sub-surface corrosion An analytical method based on Truncated Region
in layered samples. Typically, two measurement points are Eigenfunction Expansion (TREE) has been discussed in
deployed in gradient-field sensing. The two points stand at Ref. [27], which shows a good accuracy with experimental
two different z-distances to the specimen under test. test data. The use of TREE method transforms the PEC
Additionally, Joubert, et al have studied the use of mag- model from an integral into a sum of series. Fig. 8 shows
neto-optical (MO) film for imaging the structure of riveted the flowchart of the proposed method. Moreover, this
multi-layered aircraft assembly for detecting cracks that are method extends the scope of analytical method signifi-
emanating from the rivet in hidden layers of the structure cantly [41].
[33], [34]. The film is affected by the distribution of the For the measurement of the wall thickness of pipes,
magnetic fluxes induced by the eddy current and will reflect another analytical method relying on the use of inverse
the light deliberately shone on it towards a CCD camera. Laplace transform has been developed in Ref. [42]. The
In terms of the direction of the field that is measured, model is, unlike most reported analytical models, non-ax-
most probes would detect field normal to the surface of the isymmetric and it presents a good correlation with mea-
sample under test. However, there are also probes that are sured data. Rather than having an encircling coil, the
designed to detect the field parallel to the surface, such as excitation coil is perpendicular to the surface of the pipe.
the work in [35]. For complex cases, numerical algorithms like Stehfest and
Fast Fourier transform have been proposed to do the
inversion of Laplace transform, because analytical inver-
4 Modelling sion of Laplace transform is inaccessible [43]. Chen et al
also proposed new signal features that can be used for
PEC’s electromagnetic problems, just like other eddy determining the sample plate’s magnetic permeability and
current NDT&E methods, are governed by Maxwell electrical conductivity assuming that the thickness is pre-
equations. They can be solved in both time and frequency viously determined by means of their analytical model
domains. For frequency domain solutions, the time-based [44].
signals are transformed first into frequency spectrum by An analytical model based on the Fourier superposition
using Fourier transform before later, having gone through concept has been presented for modelling the crack
some processing, transformed back into the time domain in detection task in a multi-layered structure and validated by
order to get the final results. As with many other engi- using FEM results [45]. Accuracy of the model in the
neering problems, both analytical and numerical models results has been shown with fast execution time compared
are used in PEC. Analytical methods are known to be fast to the FEM solution. An example of the use of numerical
in the processing, although they are used only for relatively model is introduced in Ref. [46], which studies PEC signals
simple geometries. Reports show that researchers have when corrosion occurs in the inner wall of an insulated
attempted to build and use both analytical and numerical pipe. The signal features such as peak and zero crossing

123
506 Ali Sophian et al.

thickness, size and position of defects, and to isolate them


from the undesired parameters, such as lift-off variation.

5.1 Signal Feature and Feature Extraction

In most of PEC techniques, a reference signal is used which


is captured from a defect-free reference sample. Then a
difference signal is obtained by subtracting the reference
signal from the test one. Different types of signal are col-
lected depending whether a coil or a magnetic sensor is
used. A coil will capture the time-derivative of the mag-
netic field while a magnetic sensor will sense the field
itself. PEC time responses can be normalized in order to
reduce the effects of lift-off variation and varying magnetic
permeabilities prior to the calculation of the above time-
domain features [48]. Fig. 9 shows typical examples PEC
signals obtained by using a Hall-device.
The features that are used to infer the desired parameters
may be readily available, such as peak value [32], [48],
[49], peak arrival time [49], rising time [50] and zero
crossing time [51]. In addition to these more traditional
features, there have also been emerging features proposed
in various reports, such as the so called ‘‘relative variation
of magnetic flux’’ [20], which has been shown through both
simulation and experiments to have potentials in deter-
mining the lift-off for ferromagnetic materials [7]. This
feature is calculated by processing an interval of the decay
part of the field is used rather than the rising part and it
should not, nevertheless, be used for small lift-offs. Huang
et al discussed another feature derived from the cumulative
integration of the time signal from the decaying part [8].
Fig. 8 A TREE-based modelling method for PEC [27] The feature is related to the time constant of the first order
component of the signal and it is used to predict the
times were investigated by using the FEM model. Another
example of reported work on numerical was presented in
700
Ref. [47] that used the Fourier transform rather than time-
stepping. The validation of the model was carried out by 600
using experimental data at selected frequencies for detec-
tion of denting problems. 500
Signal intensity (arb)

400 Reference signal


5 Feature Extraction and Classification Base defect signal
300
Difference signal
PEC signals, as conventional ECT, are affected by various
200
factors, such as electrical conductivity, lift-off, magnetic
permeability, thickness of the sample and inhomogeneity 100
of the material. The other challenges faced by researchers
in obtaining useful information from the signals are the 0
0 1000 2000 3000 4000
noise and the low level signals in some cases. Conse- Time (ms)
quently, the right signal processing, signal analysis, feature
extraction and classification model must be implemented in Fig. 9 Typical PEC signals obtained by using a Hall-device-based
order to attain the desired parameters, such as coating probe

123
Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review 507

thickness of ferromagnetic plates. Another example is the the extraction than time domain features, citing the work
kurtosis coefficient, which represents the craggedness level [62] that uses spectral amplitude and phase as the signal
of a PEC response signal, that is used in sample’s edge features.
identification [52]. Finally, there are many works where both time and
In other cases, researchers try to optimise the discrimi- frequency domain features are used simultaneously. Within
nation within the range of the parameters, which require this group, time–frequency decomposition tools are often
them to employ data dimensionality reduction techniques. used, such as Wavelet, empirical mode decomposition
In some of the cases, the obtained features are subsequently (EMD) and Rihaczek distribution. Following the analysis, a
fed into a classifier in order to either classify or quantify the feature extraction technique, such as PCA and Fisher linear
defects. discriminant analysis (FLDA), is then normally used in
Dimensionality reduction techniques used in the feature order to reduce the data dimensions for defect classification
extraction for PEC signals that have been reported include purposes.
principal component analysis (PCA), independent compo- Tian, et al used wavelet transform, which captures both
nent analysis (ICA) [53] and Fisher linear discriminant temporal and spectral information from the time signals of
analysis (FLDA) [51]. PCA has been the most widely used PEC, in order to improve their classification [63]. Hosseini
since the work of Tian, et al that was reported in [54]. and Lakis use Rihaczek distribution for time–frequency
Other examples of work where PCA has been used are analysis of the signals and follow it up using PCA to reduce
introduced in Refs. [14], [28], [48], [52], [55–60]. the features, which are the used classify defects located in
The input data that are fed into the feature extraction different layers in a multi-layered specimen [60].
stage can be time domain signals, frequency domain sig- The use of both time and frequency domains is studied
nals or the combination of both domains. Examples of time by Ref. [18] for detection and characterization of cracks
domain signal input are captured in Refs. [28], [54], [14]. under fasteners in aircraft layered-structures. The combi-
In addition to the time domain features, frequency nation of the signal analysis and unique symmetric exci-
domain features are also utilized, both with and without the tation coil leads to the ability in detection of 1 mm-long
use of the time domain ones. This is understandable as PEC crack at the depth of 10 mm aided by a GMR sensor.
signals contain a wide spectrum of frequencies. And con- Although in the study, the time and frequency domain
sidering the skin depth effect, it is widely known that features were presented separately, the combination of both
higher frequencies are more sensitive to surfer or near may increase the confidence in the detection.
surface defects, while lower frequencies penetrated deeper Zhang, et al combines both time domain features,
resulting in sensitivity to more deeply buried defects. namely peak value, peak time and rising time, together
For classification of surface and sub-surface defects, He, with a frequency domain feature derived by using EMD
et al [19] used amplitudes of a few harmonics generated by and Hilbert transform for classification of surface notch,
using FFT. In another more recent work, for mapping sub-surface notches and wall losses [64]. In this work, the
artificial surface slots and holes, Abrantes, et al have also researchers do not need to use any data dimensionality
used FFT to get the amplitudes and phases of the funda- reduction.
mental and harmonics [16]. By using the amplitudes and
phases of selected frequencies, the defects can be identi- 5.2 Classifiers
fied. The phases are shown to pinpoint the centre of the
defect when they swing by 180°. The very high frequencies As mentioned earlier, in many cases researchers have
that are used should mean that only surface defects can be proposed the use of a classifier for classification or quan-
dealt with the approach. tification of defects. It is thought that this would be an
Another example of FFT-based work is done in Ref. [56], important and desirable feature for future PEC systems that
which shows the ability to discriminate different types of will potentially reduce the level of operator skill require-
defects, namely surface crack, sub-surface crack, surface ments. An automated system that will reliably provide
cavity and sub-surface cavity. An improved result is, how- information on the location and type or size of a detected
ever, shown when the time signal is decomposed first by defect would be advantageous which reduce dependency
using wavelet into the approximate and detail signals before on human interpretation skills.
being transformed into the frequency domain by using the Broadly, there are two major groups of classification
FFT. PCA is then used in order to extract the most significant methods, which are supervised and unsupervised. In
features to be used for the classification stage. supervised classification, the training data are labelled
Power spectral density analysis (PSDA) is also used by based on the class, such as defect type or location. On the
Qiu, et al [61] as frequency-domain features, which are other hand, unsupervised classification does not require the
believed by the authors to be more stable and accurate in labelling of the training data, therefore this classification is

123
508 Ali Sophian et al.

really only clustering the input data without identifying both the time and amplitude follow a unique curve against
what each cluster represents. thickness and conductivity [69]. At the early stage, LOI
Both categories of classification have been studied and phenomenon could only exist when a PEC probe using a
used in PEC NDT. The supervised classification methods coil as the receiver is placed above a nonferrous sample.
include support vector machine (SVM) and Linear Dis- Tian, et al [70] obtained LOI points by doing the derivative
criminant Analysis (LDA), whilst the used unsupervised of the output of a Hall-based probe, and Li [71] extended
include K-means and Bayes. In the majority of the reported the method to a PEC probe in which a solid-state magnetic
works, neither classifying nor clustering methods have field sensor could be placed at arbitrary positions. Further,
been used. Rather, manual human interpretation is made on Lefebvre and Mandache [72] recorded LOI phenomenon
the plots of distribution of the features. by covering a conductive but nonferrous layer on a ferro-
There are only a few reports on the use of supervised magnetic substrate. Also, Mandache and Lefebvre [73]
classification, one of them is Ref. [51], where SVM tech- made an attempt to disclose the physics of the LOI phe-
niques have been employed in conjunction with Fischer’s nomenon. It is found that the LOI phenomenon is a general
LDA (FLDA) for localization and classification of defects characteristic of the eddy currents, not only to transient
in multi-layered structure. eddy currents. Kral, et al [74] presented a linear trans-
For unsupervised methods, K-means has been used for former model to investigate the origin of LOI phenomenon,
estimation of the size the cracks, also in aircraft structures and obtained LOI points from the time derivatives of the
[28]. Another report by Qiu, et al [61] included both output of a GMR-based probe. Interestingly, a LOI point is
supervised and unsupervised methods, namely LDA and recently observed in the spectral PEC signals [67].
Bayes. Although they report both approaches managed to The use of normalisation and of two reference signals
separate the surface and sub-surface defects, from the has been proposed to reduce the effects of lift-off [75].
resulting plots it can be seen that the separation achieved Tian, et al found that lift-off invariance (LOI) existed for
by using LDA is better and has relatively wider distances magnetic sensors - not only for pick-up coils - and pro-
between the two groups of defects. posed new lift-off estimation techniques by using theoret-
ical models [70]. They believe that the techniques may also
5.3 Lift-off Problem be used when an array of pick-up sensors are used.
A technique to indirectly reduce the effects of lift-off by
Lift-off problem has always been associated with eddy using the slope of the difference signal has been proposed
current NDT, and PEC is not an exception. The detrimental [76]. The technique is potentially useful in the measure-
effect of lift-off slows down and limits the spread of the ment of depth of surface defect.
applications of PEC and ECT in NDT. Thankfully, many The work done by Huang and Wu shows that the relative
researchers have attempted to overcome this problem and rate of change of the magnetic flux is independent of lift-
proposed techniques to eliminate or reduce the effect of off and, therefore, is a good candidate for getting the
lift-off variation. Each of these techniques may only be desired information from the PEC signals, which was in the
applicable to specific PEC system implementations and inference of the thickness of ferromagnetic plates [77]. The
applications. use of normalisation in frequency domain signals can also
The most popular features to be used in order to mini- be used to reduce the effect of lift-off variation [78].
mize the effects of lift-off is called lift-off point of inter-
section (LOI), which is a point at which PEC signals
intersect when only lift-off distance varies. Therefore, LOI 6 NDT&E Applications of PEC
is inherently immune to lift-off variations, which makes a
LOI point one of desirable features to remove lift-off effect Thanks to its versatility, PEC NDT has been used in
for PEC evaluation. LOI phenomenon was originally numerous different NDT applications, both in material
observed in experiments and proposed to avoid false characterization and structural integrity inspection. In
indication due to lift-off effect when material loss in a two- material characterization, PEC has been suggested to be
layer assembly was evaluated by PEC technique [65], [66]. used for measurement of electrical conductivity and mag-
It is reported that the signals of LOI points have been netic permeability of materials. Whilst, in the structural
successfully adopted to measure the thickness of a plate integrity testing, PEC NDT has been applied for defect
[67] and quantify defects and their locations in a layered detection and characterization, evaluation of corrosion,
sample [68]. To enable the interpretation of LOI signals, measurement of insulation thickness, plate thickness and
the characteristics of LOI signals in terms of time and wall thickness of pipe. This covers both insulated and non-
amplitude were revealed under a wide variety of condi- insulated, coated and non-coated materials. Furthermore,
tions. The results demonstrate that LOI signals including still within the area of integrity testing, detection of cracks

123
Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review 509

under fasteners and between fasteners in aircraft structures materials, including copper, tungsten, inconel 600, alu-
using PEC has also been explored and implemented. Ref- minium and titanium [80].
erences to show the wide diversity of PEC applications are Röntgen Technische Dienst(RTD) – now known as
included and discussed briefly below. Applus RTD – has created a PEC system called RTD-
Incotest that is able to measure the wall thickness – ranging
6.1 Measurement of Electrical Conductivity from 6 mm up to 65 mm – of both pipes and plates made of
and Magnetic Permeability low alloy carbon steel, even when they are under insulation
of up to 150 mm in thickness [81]. By measuring the wall
The use of PEC in the measurement of electrical conduc- thickness, the corrosion under insulation (CUI) can be
tivity of ferromagnetic materials, such as carbon steel evaluated.
plates, has been investigated by Chen, et al [44]. They have Another company – Canada-based Eddyfi – has also
successfully established PEC as a non-contacting mea- made a commercial PEC system more recently. The system
surement method with a maximum error of 1.6%. In a is capable of measuring wall thickness of up to 64 mm
similar application, Desjardins, et al [10] mentioned in under non-magnetic and non-conductive insulation with
their study that the resulting conductivity from the transient thickness up to 203 mm [82]. The testing can even be
eddy current is within 0.5 MS/m from that obtained by the performed in the presence of conductive weather jackets of
four-point measurement. A carbon steel tube with wall up to 1.5 mm thick (depending on the material). The ability
thickness of 0.035’’ was used in their experiment, which of measuring wall thickness through the insulation and
was also supported by an analytical calculation with an weather jacket is clearly a unique advantage for PEC
excellent agreement. In the same work, they also showed a systems.
good result on the use of PEC in measurement of magnetic Research works related to the estimation of pipe wall
permeability. thickness include [42], [49], [83], and [84]. A potentially
effective method based on time-to-peak has been presented
6.2 Measurement of Thickness of Insulation by Xu, et al for the assessment of wall thinning of insulated
and Coating pipe [49]. The time-to-peak feature divided by the thick-
ness of a calibration sample is linearly related to wall
In ensuring its efficiency, the thickness of thermal insula- thickness within 60% of the calibration sample [49]. Such a
tion of vessels and pipes is an important parameter that feature is robust to thermal insulation and lift-off, which is
needs to be monitored regularly in petrochemical indus- verified by Park, et al [85]. In a very similar type of
tries. Experimental measurements of the thickness of application, Park et al studied the use of PEC to determine
thermal insulation on a steel plate has been done by using the wall thinning of stainless steel pipes in nuclear power
PEC and the report shows that an effective thickness plants without removing the insulation, where it could
measurement of up to 80 mm is possible [7]. estimate thickness of up to 5 mm at the lift off of 6 mm
In the area of coating thickness determination, Tai, et al [84]. Huang, et al [8] and Cheng investigated the decay
obtained excellent results for both magnetic and non- behaviour of logarithmic scaled PEC signals. It is found
magnetic coatings on both magnetic and non-magnetic that the decay coefficient is approximately linear to wall
substrates by using peak values [79]. Their work was also thickness, and is insensitive to lift-off variations [29].
supported by analytical theoretical models. Coating is very A distinctive approach by using a remote field probe has
important for corrosion protection, among others. been presented in Ref. [12] for wall thickness measurement
of ferromagnetic tubes from the inside. The researchers
6.3 Measurement of Thickness and Evaluation believe that this would be suitable for harsh-environment
of Corrosion operation.

Thickness measurement and evaluation of corrosion are 6.4 Detection and Characterization of Defects
discussed in the same section as basically the work on the
same principle and the many of the works done in thickness There is a long list of literatures covering this topic,
measurement are intended for inspection of corrosion. showing that it has been the main focus of researchers in
An example of the work in plate thickness measurement PEC. For example, detection and sizing of sub-surface
is introduced in Ref. [8], where a thickness of up to 30 mm defects in a non-magnetic specimen made of AISI type 316
was shown to be successfully measured. The plate was stainless steel has been shown in Ref. [21]. The deepest
made of Q235 steel. Another example of work in this topic notch was located 6 mm below the surface in an 8-mm-
was presented by Shin, et al who investigated different thick block. Works with similar types of artificial defects

123
510 Ali Sophian et al.

and similar shape of sample, i.e., plate or block, have also suitable and appropriate NDT inspections for ensuring safe
been reported by Tian, et al [50], [54], [63], [86]. and reliable operations with minimum downtimes.
Other examples of defect detection and characterization A reliable, yet affordable technique will obviously be
include flaws in the inner surface of pipe under insulation preferred by customers and operators alike. It is thought
[87], outer corrosion on carbon steel pipes used in oil and that the costs of the application of NDT can be reduced
gas industry [35], surface and sub-surface cracks and cavity through enhanced sensitivity and improved user-friendli-
detection in con-casting slabs (CCS) [56], subsurface cor- ness, among others. With improved sensitivities, the
rosion in a plate [4], surface corrosion on coated mild steel interval between tests can be made longer and, therefore
S275 plates [48], [88], and various types of defect on steam less tests need to be performed, which in turn, will help
generator support structures [59]. Detection of wall thin- reduce the costs. This challenge presents an opportunity for
ning in pipe structures has been included in the previous PEC, especially in the optimisation and innovation related
sub-section above. to the probe design, which should make the best use both
Aircraft industry has been given a lot of attention by the analytical and numerical modelling methods. An example
PEC research community since the very early stages of is the use a rather innovative a differential probe based on a
PEC history. Stress-induced cracks that occur under a rectangular excitation coil [94], that has been designed
fastener and between fasteners in multi-layered structures with the help of a finite element model. The probe suc-
of the wings are crucial to be detected early for the safety cessfully produces higher sensitivity in detection of defect
of the aircraft during its flights. Works related to these in different materials, including CFRP.
issues can be found in the following works about PEC [1], The sensitivity of the PEC techniques may also be
[14], [15], [18], [28], [58], [60], [89], among others. improved through the use of newer types of magnetic
Related to the aircraft industry is the new emerging sensors, such as the tunnelling magnetoresistance (TMR),
material carbon fiber reinforced plastic or CFRP for short. thanks to the continuous improvement in the sensitivity of
CFRP is growing in use in various industrial sectors thanks magnetic sensing technologies. TMR exhibits a higher
to its strength and lightness. However, its relative prone- sensitivity and higher signal-to-noise ratio (SNR) com-
ness to impact damage immediately leads to the industrial pared with hall devices and other magnetoresisitive sen-
requirement of effective NDT techniques for ensuring its sors, including GMR [95].
integrity. On the light of this, researchers have investigated As for more user-friendly systems, they would be
the application of eddy current NDT, including PEC, to depending less on the high skills of the operators for
detect and characterize defects and damage in CFRP [90], interpreting the signals and the amount of training can be
[91]. The CFRP problem is particularly challenging due to kept a minimum. All this, again, will help save the costs
its anisotropic characteristic. It can be said that the involved in in-service inspection for ensuring the integrity
achievement in this application is still limited. of structures or parts. Such systems can be materialized by
An initial study of the possibility of measuring stress in having an automated classification for defects. The use of
aluminium alloy has been reported in Ref. [92], which supervised classification techniques should be studied and
concludes that further work should be done before the developed more in order to give an answer to this chal-
technique can be realized as an inspection tool. Another lenge. Applications that may benefit from an automated
application in defect detection is one for friction stir classification functionality are those where multi types or
welding [93]. locations of defects are potentially present, which include
the inspection of aircraft multilayer structures and inspec-
tion of support structures in steam generators.
7 Challenges and Opportunities Cost reduction can also be achieved when the inspection
time can be lowered. In this case, PEC may be able to
The potential opportunities of PEC are shaped by both its benefit from the usage of multi-modal NDT and data
strengths and the progressively more quantitative require- fusion. This approach has not shown any significant
ments of NDT&E, which is characterised by ever appearance in the literature and new research can be
increasingly more stringent safety regulations; and this is undertaken to open up new opportunities for the applica-
especially true in the sectors where human lives are in tion of PEC NDT. Another way of slashing the inspection
stake, such as aerospace and nuclear industry. This would time is, as already rising in popularity with some other
make the role of NDT&E even more important and rele- NDT techniques, by the application of robotics. Again,
vant than ever. Opportunities for NDT in general and PEC opportunities in this area of research are still wide open for
in particular, are also presenting themselves as currently exploration, while there is only a handful of relevant
many structures and aircraft fleet are ageing, which need existing literature.

123
Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review 511

In manufacturing industry, there is an increasing support reported, as shown in this review. This gives approval to
for the use of NDT&E to improve the quality and consis- the theoretical understanding that PEC is rich of informa-
tency of the products. The potential ability of contactless tion that is potentially useful for quantitative NDT&E.
thickness measurement of plates or sheets should be Literature on different aspects of PEC NDT has been
exploited so that PEC with its unique benefits can be used presented and discussed. Challenges and opportunities of
in manufacturing process control for such products. The PEC NDT in the future have also been brought forward. It
ability of negating the effects of lift-off variations during is expected that this work can contribute to research work
measurement has to be achieved for generating consistently in PEC NDT, especially for researchers and developers
accurate results. If the required level of accuracy can be who are embarking a project in PEC NDT.
met, the PEC approach can be more advantageous com-
pared with, for example, the use of laser triangulation Acknowledgements The authors would like to thank Ministry of
Higher Education of Malaysia for funding the project on PEC NDT at
sensors that would need access to both sides of the plates or IIUM through the research grant FRGS16-059-0558. This work is
sheets. also supported by the National Natural Science Foundation of China
Simultaneous measurement of multiple parameters, such under research grants 51677187 and 51307172. Dr. Mengbao Fan
as conductivity, permeability and thickness is strongly would like to thank China Jiangsu Provincial Department of Educa-
tion and China University of Mining and Technology for sponsoring
demanding, and attracting increasingly more attention for his visit to Newcastle University as an academic visitor.
quality control or structural integrity in some cases. In
practice, multi-sensing modalities are performed for a
comprehensive assessment, which is costly, labour-intensive
and sometimes time-consuming. Fortunately, PEC is thought References
to be a feasible and efficient tool to measure conductivity,
permeability and thickness simultaneously. Chen, et al [44] 1. Lebrun B, Jayet Y, Baboux J-C. Pulsed eddy current signal
analysis: application to the experimental detection and charac-
proposed an analytical-model-based method to infer con- terization of deep flaws in highly conductive materials. NDT E
ductivity and permeability with the prerequisite of knowing Int., 1997, 30(3): 163–170.
the wall thickness beforehand. Adewale, et al [96] investi- 2. Michniakova M, Janousek L. The impact of various waveform
gated the contribution of permeability and conductivity on shapes on the response signal in pulsed eddy current NDE.
ELEKTRO 2012, 2012: 439–442.
PEC signals. It is found that conductivity effects are 3. Li J, Wu X, Zhang Q, et al. Pulsed eddy current testing of fer-
prominent in the rising edge of the transient response, romagnetic specimen based on variable pulse width excitation.
whereas permeability effects dominate in the steady stage. NDT E Int., 2015, 69: 28–34.
This finding presents a feasible way to decouple permeability 4. Li Y, Yan B, Li D, et al. Pulse-modulation eddy current
inspection of subsurface corrosion in conductive structures. NDT
and conductivity, and implies a feasible method to infer these E Int., 2016, 79: 142–149.
two parameters within one measurement. Desjardins, et al 5. Abidin I Z, Mandache C, Tian G Y, et al. Pulsed eddy current
[10] created a novel forward-solutions-based inverse method testing with variable duty cycle on rivet joints. NDT E Int., 2009,
for simultaneous and accurate characterization of perme- 42(7): 599–605.
6. Dolabdjian C P, Perez L, De Haan V O, et al. Performance of
ability and conductivity by computing the areas under scaled magnetic pulsed-eddy-current system using high dynamic and
PEC signals. At present, the developed methods are only high linearity improved giant magnetoresistance magnetometer.
proven to be feasible. However, much work still needs to be IEEE Sensors Journal, 2006, 6(6): 1511–1517.
done to make these methods applicable. For example, the 7. Li J, Wu X, Zhang Q, et al. Measurement of lift-off using the
relative variation of magnetic flux in pulsed eddy current testing.
developed methods need to be extended in order to be able to NDT E Int., 2015, 75: 57–64.
determine four or more parameters or be applied to a layered 8. Huang C, Wu X. An improved ferromagnetic material pulsed
sample. Further, the accuracy of the existing methods greatly eddy current testing signal processing method based on numerical
relies on the correlation between calculated and experi- cumulative integration. NDT E Int., 2015, 69: 35–39.
9. Blitz J. Eddy current testing of metails. NDT E Int., 1997, 30(1):
mental data. The efforts to reduce the discrepancy should be 41.
made for accuracy improvement. In addition, uncertainty 10. Desjardins D, Krause T W, Clapham L. Transient eddy current
analysis should also be performed to evaluate the proposed method for the characterization of magnetic permeability and
methods [97]. conductivity. NDT E Int., 2016, 80: 65–70.
11. Desjardins D R, Krause T W, Clapham L. Transient response of a
driver coil in transient eddy current testing. NDT E Int., 2015, 73:
8–14.
8 Conclusions 12. Vasic D, Bilas V, Ambrus D. Pulsed eddy-current nondestructive
testing of ferromagnetic tubes. IEEE Transactions on Instru-
mentation and Measurement, 2004, 53(4): 1289–1294.
PEC NDT&E has received a great deal of attention from 13. Park D G, Angani C S, Cheong Y M. Differential pulsed eddy
researchers in the past 30 years and more, and yet, new current probe to detect the sub surface cracks in a stainless steel
applications and techniques are still being proposed and pipe. 18th World Conference on Nondestructive Testing, 2012.

123
512 Ali Sophian et al.

14. Horan P, Underhill P R, Krause T W. Pulsed eddy current 36. Dai X W, Ludwig R, Palanisamy R. Numerical simulation of
detection of cracks in F/A-18 inner wing spar without wing skin pulsed eddy-current nondestructive testing phenomena. IEEE
removal using Modified Principal Component Analysis. NDT E Transactions on Magnetics, 1990: 26(6): 3089–3096.
Int., 2013, 55: 21–27. 37. Ludwig R, Dai X W. Numerical and analytical modeling of
15. He Y, Luo F, Pan M, et al. Pulsed eddy current technique for pulsed eddy currents in a conducting half-space. IEEE Transac-
defect detection in aircraft riveted structures. NDT E Int., 2010, tions on Magnetics, 1990: 26(1): 299–307.
43(2): 176–181. 38. Allen B L, Lord W. Finite element modeling of pulsed eddy
16. Abrantes R F, Rosado L S, Piedade M, et al. Pulsed eddy currents current phenomena. Review of Progress in Quantitative Nonde-
testing using a planar matrix probe. Measurement, 2016, 77: structive Evaluation, 1984: 3(A): 561–568.
351–361. 39. Allen B, Ida N, Lord W. Finite element modeling of pulse Eddy
17. Southwest Research Institute. Southwest Research Institute 2004 current NDT phenomena. IEEE Transactions on Magnetics,
Annual Report, 2004. 1985, 21(6): 2250–2253.
18. Yang G, Tamburrino A, Udpa L, et al. Pulsed eddy-current based 40. Desjardins D P R, Krause T W, Gauthier N. Analytical modeling
giant magnetoresistive system for the Inspection of aircraft of the transient response of a coil encircling a ferromagnetic
structures. IEEE Transactions on Magnetics, 2010, 46(3): conducting rod in pulsed eddy current testing. NDT E Int., 2013,
910–917. 60: 127–131.
19. He Y, Pan M, Luo F, et al. Pulsed eddy current imaging and 41. Fan M, Huang P, Ye B, et al. Analytical modeling for transient
frequency spectrum analysis for hidden defect nondestructive probe response in pulsed eddy current testing. NDT E Int., 2009,
testing and evaluation. NDT E Int., 2011, 44(4): 344–352. 42(5): 376–383.
20. He Y, Luo F, Pan M, et al. Defect classification based on rect- 42. Chen X, Lei Y. Time-domain analytical solutions to pulsed eddy
angular pulsed eddy current sensor in different directions. Sensors current field excited by a probe coil outside a conducting ferro-
Actuators A Phys., 2010, 157(1): 26–31. magnetic pipe. NDT E Int., 2014, 68: 22–27.
21. Arjun V, Sasi B, Rao B P C, et al. Optimisation of pulsed eddy 43. Fan M, Cao B, Wang Y. Computation of coil-induced voltage due
current probe for detection of sub-surface defects in stainless steel to a defect-free plate using Stehfest’s method for pulsed eddy
plates. Sensors Actuators A Phys., 2015, 226: 69–75. current evaluation. Insight Non-Destructive Test. Cond. Monit.,
22. Zhou D, Wang J, He Y, et al. Influence of metallic shields on 2010, 52(6): 302–304.
pulsed eddy current sensor for ferromagnetic materials defect 44. Chen X, Lei Y. Electrical conductivity measurement of ferro-
detection. Sensors Actuators A Phys., 2016, 248: 162–172. magnetic metallic materials using pulsed eddy current method.
23. Fan M, Cao B, Sunny A I, et al. Pulsed eddy current thickness NDT E Int., 2015, 75: 33–38.
measurement using phase features immune to liftoff effect. NDT 45. Theodoulidis T, Wang H, Tian G Y. Extension of a model for
E Int., 2017, 86: 123–131. eddy current inspection of cracks to pulsed excitations. NDT E
24. Tumanski S. Induction coil sensors—a review. Meas. Sci. Tech- Int., 2012, 47: 144–149.
nol., 2007, 18(3): R31–R46. 46. Liu S X, Xin W, Ding K Q. Simulation of corrosion on detection
25. Ripka P, Janosek M. Advances in magnetic field sensors. IEEE for pulsed eddy current. Fuzzy Systems and Knowledge Discovery
Sensors Journal, 2010, 10(6): 1108–1116. (FSKD), 2010 Seventh International Conference on, 2010, 4:
26. Herrera-May A L, Aguilera-Cortés L A, Garcı́a-Ramı́rez P J, 1839–1842.
et al. Resonant magnetic field sensors based on MEMS technol- 47. Pavo J. Numerical calculation method for pulsed eddy-current
ogy. Sensors, 2009, 9(10): 7785. testing. IEEE Transactions on Magnetics, 2002, 38(2):
27. Li Y, Tian G Y, Simm A. Fast analytical modelling for pulsed 1169–1172.
eddy current evaluation. NDT E Int., 2008, 41(6): 477–483. 48. Alamin M, Tian G Y, Andrews A, et al. Principal component
28. Kim J, Yang G, Udpa L, et al. Classification of pulsed eddy analysis of pulsed eddy current response from corrosion in mild
current GMR data on aircraft structures,’’ NDT E Int., 2010, steel. IEEE Sensors Journal, 2012, 12(8): 2548–2553.
43(2): 141–144. 49. Xu Z, Wu X, Li J, et al. Assessment of wall thinning in insulated
29. Cheng W. Pulsed eddy current testing of carbon steel pipes’ wall- ferromagnetic pipes using the time-to-peak of differential pulsed
thinning through insulation and cladding. J. Nondestruct. Eval., eddy-current testing signals. NDT E Int., 2012, 51: 24–29.
2012, 31(3): 215–224. 50. Tian G Y, Sophian A. Defect classification using a new feature
30. Albrecht J, Carls S, Robert S. Magnetoresistive Sensors for for pulsed eddy current sensors. NDT & E International, 2005,
Nondestructive Evaluation. 10th SPIE International Symposium, 38(1): 77–82.
Nondestructive Evaluation for Health Monitoring and Diagnos- 51. Chen X, Hou D, Zhao L, et al. Study on defect classification in
tics, Conference 5770, 2005. multi-layer structures based on Fisher linear discriminate analysis
31. Panaitov G, Krause H-J, Zhang Y. Pulsed eddy current transient by using pulsed eddy current technique. NDT E Int., 2014, 67:
technique with HTS SQUID magnetometer for non-destructive 46–54.
evaluation. Phys. C Supercond., 2002, 372–376: 278–281. 52. BinFeng Y, FeiLu L, Dan H. Research on edge identification of a
32. Li Y, Yan B, Li D, et al. Gradient-field pulsed eddy current defect using pulsed eddy current based on principal component
probes for imaging of hidden corrosion in conductive structures. analysis. NDT E Int., 2007, 40(4): 294–299.
Sensors Actuators A Phys., 2016, 238: 251–265. 53. Yang G, Tian G Y, Que P W, et al. Independent component
33. Joubert P Y, Le Diraison Y, Xi Z, et al. Pulsed eddy current analysis–based feature extraction technique for defect classifica-
imaging device for non destructive evaluation applications. tion applied for pulsed eddy current NDE. Res. Nondestruct.
Sensors, 2013 IEEE, 2013: 1–4. Eval., 2009, 20(4): 230–245.
34. Joubert P Y, Le Diraison Y. Pulsed eddy current imager for the 54. Sophian A, Tian G Y, Taylor D, et al. A feature extraction
enhanced non destructive evaluation of aeronautical riveted technique based on principal component analysis for pulsed Eddy
assemblies. IEEE SENSORS 2014 Proceedings, 2014: current NDT. NDT & E International, 2003, 36(1): 37–41.
1188–1191. 55. Horan P F, Underhill P R, Krause T W. Real time pulsed eddy
35. Majidnia S, Nilavalan R, Rudlin J. Investigation of an encircling current detection of cracks in F/A-18 inner wing spar using dis-
Pulsed Eddy Current probe for corrosion detection. IEEE Sensors criminant separation of modified principal components analysis
2014 Proceedings, 2014: 835–838. scores. IEEE Sensors Journal, 2014, 14(1): 171–177.

123
Pulsed Eddy Current Non-destructive Testing and Evaluation: A Review 513

56. Qiu X, Zhang P, Wei J, et al. Defect classification by pulsed eddy 77. Huang C, Wu X. Probe lift-off compensation method for pulsed
current technique in con-casting slabs based on spectrum analysis eddy current thickness measurement. Antennas and Propagation
and wavelet decomposition. Sensors Actuators A Phys., 2013, (APCAP), 2014 3rd Asia-Pacific Conference on., 2014: 937–939.
203: 272–281. 78. Tian G Y, He Y, Adewale I, et al. Research on spectral response
57. Pan M, He Y, Tian G, et al. Defect characterisation using pulsed of pulsed eddy current and NDE applications. Sensors Actuators
eddy current thermography under transmission mode and NDT A Phys., 2013, 189: 313–320.
applications. NDT E Int., 2012, 52: 28–36. 79. Yang H-C, Tai C-C. Pulsed eddy-current measurement of a
58. Stott C A, Underhill P R, Babbar V K, et al. Pulsed eddy current conducting coating on a magnetic metal plate. Meas. Sci. Tech-
detection of cracks in multilayer aluminum lap loints. IEEE nol., 2002, 13(8): 1259.
Sensors Journal, 2015, 15(2): 956–962. 80. Shin Y-K, Choi D-M, Kim Y-J, et al. Signal characteristics of
59. Buck J A, Underhill P R, Mokros S G, et al. Pulsed eddy current differential-pulsed eddy current sensors in the evaluation of plate
inspection of support structures in steam generators. IEEE Sen- thickness. NDT E Int., 2009, 42(3): 215–221.
sors Journal, 2015, 15(8): 4305–4312. 81. Robers M A, Scottini R. Pulsed eddy currents in corrosion
60. Hosseini S, Lakis A A. Application of time–frequency analysis detection. NDT.net, 2002: 7(10).
for automatic hidden corrosion detection in a multilayer alu- 82. Eddyfi. Lyft Specifications Sheet. [Online]. Available: http://
minum structure using pulsed eddy current. NDT E Int., 2012, 47: www.eddyfi.com/wp-content/uploads/2016/09/specifications-
70–79. sheet-lyft.pdf. [Accessed: 02-Oct-2016].
61. Qiu X B, Liu L L, Li C L, et al. Defect classification by pulsed 83. Xie S, Chen Z, Takagi T, et al. Quantitative non-destructive
eddy-current technique based on power spectral density analysis evaluation of wall thinning defect in double-layer pipe of nuclear
combined with wavelet transform. IEEE Trans. Magn., 2014, power plants using pulsed ECT method. NDT E Int., 2015, 75:
50(9): 1–8. 87–95.
62. Kiwa T, Kawata T, Yamada H, et al. Fourier-transformed eddy 84. Park D G, Angani C S, Kim G D, et al. Evaluation of pulsed eddy
current technique to visualize cross-sections of conductive current response and detection of the thickness variation in the
materials. NDT E Int., 2007, 40(5): 363–367. stainless steel. IEEE Transactions on Magnetics, 2009, 45(10):
63. Tian G Y, Sophian A, Taylor D, et al. Wavelet-based PCA defect 3893–3896.
classification and quantification for pulsed eddy current NDT. 85. Park D-G, Angani C S, Kishore M B, et al. The effects of lift-off
IEE Proceedings - Science, Measurement and Technology, 2005, from wall thinning signal in pulsed eddy current testing. J.
152(4): 141–148. Magn., 2012, 17(4): 298–301.
64. Zhang Q, Chen T, Yang G, et al. Time and frequency domain 86. Chen T, Tian G Y, Sophian A, et al. Feature extraction and
feature fusion for defect classification based on pulsed eddy selection for defect classification of pulsed eddy current NDT.
current NDT. Res. Nondestruct. Eval., 2012, 23(3): 171–182. NDT E Int., 2008, 41(6): 467–476.
65. Giguère S, Lepine B A, Dubois J M S. Pulsed eddy current 87. Lai S, Chen D Y, Chen H, et al. Pulsed eddy current testing of
technology: characterizing material loss with gap and lift-off inner wall flaws in pipe under insulation. Procedia Eng., 130:
variations. Res. Nondestruct. Eval., 2001, 13(3): 119–129. 1658–1664.
66. Giguère S, Dubois S J M. Pulsed eddy current: Finding corrosion 88. Tian G, He Y, Alamin M, et al. Corrosion characterisation using
independently of transducer lift-off. AIP Conf. Proc., 2000, pulsed eddy current sensor systems. Sensing Technology (ICST),
509(1): 449–456. 2011 Fifth International Conference on., 2011: 498–503.
67. Angani C S, Ramos H G, Ribeiro A L, et al. Lift-off point of 89. Sophian A, Tian G Y, Taylor D, et al. Design of a pulsed eddy
intersection feature in transient eddy-current oscillations method current sensor for detection of defects in aircraft lap-joints.
to detect thickness variation in stainless steel. IEEE Transactions Sensors Actuators, A Phys., 2002: 101(1–2): 92–98.
on Magnetics, 2016, 52(6): 1–8. 90. Wu J, Zhou D, Wang J, et al. Surface crack detection for carbon
68. Liu Z, Ramuhalli P, Safizadeh S, et al. Combining multiple fiber reinforced plastic (CFRP) materials using pulsed eddy cur-
nondestructive inspection images with a generalized additive rent testing. Nondestructive Evaluation/Testing (FENDT), 2014
model. Meas. Sci. Technol., 2008, 19(8): 817–822. IEEE Far East Forum on., 2014: 181–185.
69. Lefebvre J H V, Dubois J M S. Lift-off point of intercept (LOI) 91. He Y, Tian G, Pan M, et al. Non-destructive testing of low-energy
behavior. AIP Conference Proceedings, 2005, 760: 523–530. impact in CFRP laminates and interior defects in honeycomb
70. Tian G Y, Li Y, Mandache C. Study of lift-off invariance for sandwich using scanning pulsed eddy current. Compos. Part B
pulsed eddy-current signals. IEEE Trans. Magn., 2009, 45(1): Eng., 59: 196–203.
184–191. 92. Morozov M, Tian G Yun, Withers P J. The pulsed eddy current
71. Li Y B, Chen Z B, Qi Y. Generalized analytical expressions of response to applied loading of various aluminium alloys. NDT E
liftoff intersection in PEC and a liftoff-intersection-based fast Int., 2010, 43(6): 493–500.
inverse model. IEEE Trans. Magn., 2011, 47(10): 2931–2934. 93. Rosado L S, Santos T G, Ramos P M, et al. A differential planar
72. Lefebvre J H V, Mandache C. Pulsed eddy current thickness eddy currents probe: Fundamentals, modeling and experimental
measurement of conductive layers over ferromagnetic substrates. evaluation. NDT E Int., 2012: 51: 85–93.
Int. J. Appl. Electromagn. Mech., 2008, 27(1–2): 1–8. 94. Zhou D, Wang J, Wu J, et al. Investigation of rectangular dif-
73. Mandache C, Lefebvre J H V. Transient and harmonic eddy cur- ferential probes for pulsed eddy current non-destructive testing.
rents: Lift-off point of intersection. NDT E Int., 2006, 39(1): 57–60. Insight - Non-Destructive Test. Cond. Monit., 2016, 58(2):
74. Kral J, Smid R, Ramos H M G, et al. The lift-off effect in Eddy 87–100.
currents on thickness modeling and measurement. IEEE Trans. 95. Sergeeva-Chollet N, Decitre J, Fermon C, et al. Intrinsic noise
Instrum. Meas., 2013, 62(7): 2043–2049. study of Eddy current probes based on magnetoresistive sensors.
75. Tian G Y, Sophian A. Reduction of lift-off effects for pulsed eddy 11th European Conference on Non-Destructive Testing (ECNDT
current NDT. NDT E Int., 2005, 38(4): 319–324. 2014), 2014.
76. Yu Y, Yan Y, Wang F, et al. An approach to reduce lift-off noise 96. Adewale I D, Tian G Y. Decoupling the influence of permeability
in pulsed eddy current nondestructive technology. NDT E Int., and conductivity in pulsed eddy-current measurements. IEEE
2014, 63: 1–6. Transactions on Magnetics, 2013, 49(3): 1119–1127.

123
514 Ali Sophian et al.

97. Fan M, Cao B, Yang P, et al. Elimination of liftoff effect using a He has coordinated several research projects from the Engineering
model-based method for eddy current characterization of a plate. and Physical Sciences research Council (EPSRC), Royal Academy of
NDT E Int., 2015, 74: 66–71. Engineering and FP7. Also he has good collaboration with leading
industrial companies, such as Airbus, Rolls Royce, BP, nPower,
Network Railand TWI among others. He has published more than 300
Ali Sophian, born in 1974 in Indonesia, is currently an Assistant papers in Journals and peer-reviewed conferences with h-index 42.
Professor with Mechatronics Engineering Department at the Faculty
of Engineering, International Islamic University Malaysia, Gombak, Mengbao Fan, was born in Shandong China, in 1981. He received his
Malaysia. He received his B.Eng and PhD in Electronics Engineering B.Sc. degree in automation from the China University of Petroleum,
from the University of Huddersfield, UK in 1998 and 2003 China, and Ph.D degree in control science and engineering from
respectively. He has published more than 20 papers, mainly in Zhejiang University, China, in 2004 and 2009, respectively. Since
electromagnetic NDT as the outcomes of the research projects funded 2009 to 2011, he was a Lecturer with the School of Mechatronic
by the TWI Ltd, EPSRC (UK) and Universities UK. Currently, he is Engineering, China University of Mining and Technology. Since
leading research projects in non-destructive testing (NDT), Sensors 2012, He has been an associate professor with the School of
and Image Processing funded mainly by Malaysia’s Ministry of Mechatronic Engineering, China University of Mining and Technol-
Higher Education (MOHE). ogy. From January 2015 to January 2016, he academically visited the
School of Electrical and Electronic Engineering, Newcastle Univer-
Guiyun Tian, born in 1965, received his B.Sc. degree and M.Sc. sity, Newcastle upon Tyne, UK. His current research interests include
degreefrom University of Sichuan, Chengdu, China in 1985 and 1988, nondestructive testing and evaluation for quality control and struc-
respectively, and Ph.D. from University of Derby, Derby, UK, in tural integrity, modeling of eddy current field and computation,
1998. He is currently the Professor of Sensor Technologies at School sensors and instrumentation. He is an author of more than 40 technical
of Electrical and Electronic Engineering, Newcastle University, UK. papers on journals and conference proceedings. He has coordinated
He is also adjunct professor in the School of Automation Engineering, multiple research projects, such as National Natural Science Foun-
University of Electronics Science and Technology of China. His dation of China, and Jiangsu Provincial Natural Science Foundation
research interests include electromagnetic sensors, sensor array and of China, and also has good collaboration with industries.
sensor network, electromagnetic non-destructive evaluation,
advanced signal processing and integrative systems and applications.

123

View publication stats

You might also like