Pulsed Eddy Current Testing Signals Research Articles

Enhancing corrosion detection in pulsed eddy current testing systems through autoencoder-based unsupervised learning

Pulsed Eddy Current Testing (PECT) stands out as an advanced method in Non-Destructive Testing due to its extensive spectrum characteristics in comparison to traditional ECT techniques, making it exceptionally suitable for identifying corrosion. Nonetheless, the analysis of PECT signals for corrosion detection poses a challenge due to the transient nature of these signals and the impact of sensor lift-off effects. As a result, conventional methods are facing hurdles in dealing with corrosion signals of poor quality. In this study, the challenge is addressed by employing unsupervised learning methods utilizing an autoencoder neural network. This autoencoder integrates Long Short-Term Memory and 1D convolutional layers, acquiring the underlying features of normal PECT signals from non-corrosive regions. Significantly, the model is trained exclusively on this normal data, thereby obviating the necessity for pre-existing corrosion information. Through learning the inherent structure of normal signals, the model can detect anomalies in unseen data, potentially indicating corrosion. The unsupervised framework presents several advantages, such as reducing reliance on prior corrosion knowledge, mitigating inherent noise, and addressing sensor lift-off effects. Experimental results were conducted to compare with traditional methods like the lift-off of intersection and lift-off compensation methods. This approach resulted in a significant improvement in SNR, ranging from 100 % to 200 %, thus facilitating more robust NDT applications employing smart PECT sensors empowered by unsupervised learning techniques.

Inspection of Liner Wall Thinning and Interface Debonding in Bimetallic Lined Pipes Using Pulsed Eddy Current Testing.

Bimetallic lined pipe (BLP) has been increasingly used in offshore and subsea oil and gas structures, but how to identify the invisible inner defects such as liner wall thinning and interface debonding is a challenge for future development. A nondestructive testing (NDT) method based on pulsed eddy current testing (PECT) has been proposed to face these difficulties. The inspection of the BLP specimen (AISI1020 base tube and SS304 liner) is implemented from outside of the pipe by using a transmitter-receiver-type PECT probe consisting of two induction coils. By simplifying the BLP specimen to stratified conductive plates, the electromagnetic field interaction between the PECT probe and specimen is analytically modeled, and the probe inspection signals due to liner wall thinning and interface debonding are calculated. In order to highlight the weak response (in microvolts) from the liner, the inspection signals are subtracted by the signal, which is calculated in the case of only having a base tube, yielding differential PECT signals. The peak voltage of the differential signal is selected to characterize the liner wall thinning and interface debonding due to its distinguishable and linear variation. Experiment verification is also carried out on a double-walled specimen simulated by a combination of a Q235 casing pipe and SS304 tubes of different sizes. The experimental results basically agree with the analytical predictions. The peak value of the PECT signal has an ascending and descending variation with the increase in the remaining liner wall thickness and debonding gap, respectively, while the negative peak value shows opposite changes. The peak value exhibits a larger sensitivity than the negative peak value. The proposed method shows potential promise in practical applications for the evaluation of the inner defects in BLP lines.

Detection of brazing defects in stainless steel core plate using the first peak value of pulsed eddy current testing signals

As a novel building structure element, stainless steel core plate (SSCP) has seen growing application in modular construction. It is composed of two stainless steel skin panels and a number of stainless steel core tubes, which are brazed by elevated temperature inert gas. The underlying defects in brazing joints severely damage the SSCP’s structural integrity and weaken its mechanical properties. In this study, a nondestructive testing (NDT) method based on pulsed eddy current testing (PECT) was employed to contactlessly detect the missing solder and non-bonding brazing defects from the plate surface. The defect-free and defective brazing joints are simplified to stratified structures and analytical modeling for a PECT probe over the brazing joint is thus performed. The time-domain features of the calculated PECT signal due to the structure parameter variation are investigated. The first peak value is selected to characterize the brazing quality due to its distinct variation w.r.t. well-bonded brazing, missing solder, non-bonding and skin panel. Experiments are also carried out to validate the analytical modeling. With the aid of an industrial robotic arm, one-dimensional (1D) line-scan and two-dimensional (2D) planar scan of the PECT probe over the SSCP specimen are implemented and the detection results are represented in images. In contrast to the well-bonded joint, the brazing joints with missing solder defects are imaged as notched rings, on which the notch orientations and sizes agree with the actual openings of the brazing foils, while the non-bonding brazing joint is imaged as a full ring with faded color. The proposed method shows a promising aspect in practical application for automated NDT of the brazing defects in SSCPs.

Parameter Optimisation of Wavelet Denoising for Pulsed Eddy Current Signals Based on Particle Swarm Optimisation Algorithm

ABSTRACT To ensure the reliable operation of high voltage cables, it is crucial to routinely inspect the lead sealing using pulsed eddy current. But nevertheless, the signals can be impacted by noise in actual engineering. While retaining the useful high-frequency information in the original signal, wavelet denoising can filter the clutter efficiently. However, previous research on wavelet denoising has only used it as a straightforward filtering tool, ignoring the impact of changing the parameters on the actual denoising effect. In this study, a more accurate evaluation index called P NCC is constructed for pulsed eddy current testing signals with a focus on maintaining the peak information. This index is then adopted as a fitness function for the particle swarm optimisation algorithm to determine the befitting wavelet denoising parameters and denoised signals. The results show that the signal-to-noise ratio of the signals after denoising is approximately 25 dB, and the distortion at the peak position is stable at about 1%. It demonstrates that the comprehensive index P NCC is a reliable metric for assessing the quality of pulsed eddy current signals. The optimal wavelet denoising parameters for pulsed eddy current signals are found to be the Sym 4 wavelet, 10-layer decomposition and Median threshold function.

Study on the Effect of Metal Mesh on Pulsed Eddy-Current Testing of Corrosion under Insulation Using an Early-Phase Signal Feature

Corrosion under insulation (CUI) is a major threat to the structural integrity of insulated pipes and vessels. Pulsed eddy-current testing (PECT) is well known in the industry for detecting CUI, but its readings can be easily influenced by nearby conductive objects, including the insulation supporting metal mesh. As a sequel to our previous study, this paper focuses on the surface distribution of eddy currents at the time of the turning off of the driving voltage instead of examining the overall process of eddy current diffusion. Based on the fact that CUI takes place on the outside of the insulated specimen, the probe footprint was calculated only on the specimen surface. The corrosion depth was regarded as an increment to the probe lift-off, whose information was carried in the early PECT signal. Finite element simulations were performed to facilitate the calculation of the probe footprint and predict the signal behavior. The peak value, which appeared in the early phase of the differential PECT signal, was found to be well correlated with the corrosion depth. Further studies revealed that the mild steel mesh could result in the enlargement of the probe footprint and a decrease in the change rate of the peak value in relation to the corrosion depth. Finally, experiments were conducted to verify the simulation results. The presented findings are consistent with the previously reported results and provide a potential alternative to evaluate CUI in specific scenarios where the insulation has a fixed and uniform thickness.

Intelligent Hardness Prediction of Bearing Rings Using Pulsed Eddy Current Testing

Bearings are central components of rotating machinery. Original equipment manufacturers must utilize high-quality bearings for desirable performances and safety of the equipment. Hardness, which is directly linked to product performance and reliability, is, thus, an essential mechanical property of bearings. As a result, it is necessary to evaluate the hardness during the manufacturing process. However, because of the complicated transformation of microstructures during heat treatment, there is no an established noncontact, quick, and affordable method for measuring the hardness of bearings. This article describes a pulsed eddy current testing (PECT)-based novel method utilizing deep learning for reliable nondestructive testing of bearing rings. The continuous wavelet transform (CWT) was used to combine the time-frequency information in the PECT signals, and the deep learning models were trained to predict hardness. In addition, we compared and optimized the network’s performance in feature maps, the number of convolutional kernels, and the learning rate. The results indicate that the developed novel method has an average error of 1.37% and a maximum error of 1.89%, exhibiting good hardness prediction accuracy and stability.

Nondestructive Testing of Local Incomplete Brazing Defect in Stainless Steel Core Panel Using Pulsed Eddy Current.

Stainless steel core panel is a novel structure for fast modular building, but its brazing foils are susceptible to defects due to the difficulty of precisely controlling the brazing process. An automated, nondestructive testing technique is highly desirable for quick inspection of the brazing defects buried in the stainless-steel core panel. In this paper, pulsed eddy current testing (PECT) was employed to inspect local incomplete brazing defects. Finite element simulation and experiment verification were conducted to investigate the feasibility and effectiveness of the proposed method. The peak value of the PECT signal was found to be sensitive to the presence of the defect. With the aid of an industrial robotic arm, line and two-dimensional scans were performed of the PECT probe above the panel specimen. The prefabricated incomplete brazing foil was successfully imaged as a notched ring, whose opening coincides with the physical length of the missing brazing. The proposed method shows potential to serve as an effective tool for in-line or off-line automated nondestructive testing of the brazing defects in stainless steel core panels.

An analytical solution for vertical coils near a multi-layered metallic pipe in Pulsed Eddy Current Testing

Analytical model for Pulsed Eddy Current Testing (PECT) of a multi-layered metallic pipe has not been solved, which limits the development of PECT technology. In this paper, the analytical model for vertical coils near a multi-layered metallic pipe is proposed. However, the analytical solution is hard to solve due to the large number of undetermined coefficients. Combined with the global matrix method, an improved transfer matrix across a conductor is established, which overcomes unsolvable problem in the traditional transfer matrix method. The induced voltage of pickup coil calculated based on the analytical model is verified by Finite Element Method (FEM) and experiments. Furthermore, the influence of several working conditions of pipelines on the PECT signal is studied based on the analytical model, providing guidance for engineering applications.

Flaw characterization in conductive media based on pulsed Eddy current measurements: A fast non‐iterative inversion approach

This paper proposes a non-iterative procedure for flaw(s) characterization based on Pulsed Eddy Current Testing (PECT) signals analysis. The adopted inversion strategy is based on the use of supervised statistical learning algorithms. A numerical forward solver, based on the Finite Integration Technique (FIT), is used for the generation of the training data (the input-targets couples of the learning algorithm). Predictions are then carried out in almost real-time using a non-linear kernel based regression method, known as kernel ridge regression. It turns out that the direct fit of the regression model to the raw PECT signals may lead to poor prediction accuracy due to the large cardinality of PECT signals. To remedy this problem, an adaptive sampling strategy has been adopted in this work. The performance of the proposed methodology is discussed and compared with solutions proposed in the literature.

Intelligent Defect Identification Based on PECT Signals and an Optimized Two-Dimensional Deep Convolutional Network

Accurate and rapid defect identification based on pulsed eddy current testing (PECT) plays an important role in the structural integrity and health monitoring (SIHM) of in-service equipment in the renewable energy system. However, in conventional data-driven defect identification methods, the signal feature extraction is time consuming and requires expert experience. To avoid the difficulty of manual feature extraction and overcome the shortcomings of the classic deep convolutional network (DCNN), such as large memory and high computational cost, an intelligent defect recognition pipeline based on the general Warblet transform (GWT) method and optimized two-dimensional (2-D) DCNN is proposed. The GWT method is used to convert the one-dimensional (1-D) PECT signal to a 2D grayscale image used as the input of 2D DCNN. A compound method is proposed to optimize the baseline VGG16, a well-known DCNN, from four aspects including reducing the input size, adding batch normalization layer (BN) after every convolutional layer(Conv) and fully connection layer (FC), simplifying the FCs, and removing unimportant filters in Convs so as to reduce memory and computational costs while improving accuracy. Through a pulsed eddy current testing (PECT) experiment considering interference factors including liftoff and noise, the following conclusion can be obtained. The time-frequency representation (TFR) obtained by the GWT method not only has excellent ability in terms of the transient component analysis but also is less affected by the reduction of image size; the proposed optimized DCNN can accurately identify defect types without manual feature extraction. And compared to the baseline VGG16, the accuracy obtained by the optimized DCNN is improved by 7%, to about 99.58%, and the memory and computational cost are reduced by 98%. Moreover, compared with other well-known DCNNs, such as GoogLeNet, Inception V3, ResNet50, and AlexNet, the optimized network has significant advantages in terms of accuracy and computational cost, too.

A hybrid nondestructive testing method of pulsed eddy current testing and electromagnetic acoustic transducer techniques for simultaneous surface and volumetric defects inspection

In this paper, a hybrid nondestructive testing (NDT) method combining with the pulsed eddy current testing (PECT) method and the electromagnetic acoustic transducer (EMAT) method has been proposed and validated through numerical simulations and experiments. First, a numerical code is developed for the simulation of hybrid EMAT/PECT signals based on the developed EMAT and PECT code. Second, based on the numerical simulation, the influences of the eddy current induced by the excitation coil and the eddy current due to the velocity effect of the ultrasonic wave are compared and analyzed. In addition, the features of the EMAT and the PECT signals are analyzed respectively. Third, several signal separation and extraction methods are proposed on the basis of the spectral analysis and filtering strategies for extracting EMAT and PECT signals from the mixed signals of the hybrid EMAT/PECT method and their validity are evaluated through experiments. Finally, hybrid EMAT/PECT experiments are conducted and three types of defects (surface defects, bottom thinning defects, composite defects) in an aluminum plate and a SUS304 plate are successfully detected at the same time using the proposed hybrid NDT method and the signal feature parameters. Based on the numerical and experimental results, the proposed hybrid PECT/EMAT NDT method is demonstrated both high detectability and high efficiency for detecting surface and volumetric defects at the same time.

Pulsed eddy current signal denoising based on singular value decomposition

The noise as an undesired phenomenon often appears in the pulsed eddy current testing (PECT) signal, and it is difficult to recognize the character of the testing signal. One of the most common noises presented in the PECT signal is the Gaussian noise, since it is caused by the testing environment. A new denoising approach based on singular value decomposition (SVD) is proposed in this paper to reduce the Gaussian noise of PECT signal. The approach first discusses the relationship between signal to noise ratio (SNR) and negentropy of PECT signal. Then the Hankel matrix of PECT signal is constructed for noise reduction, and the matrix is divided into noise subspace and signal subspace by a singular valve threshold. Based on the theory of negentropy, the optimal matrix dimension and threshold are chosen to improve the performance of denoising. The denoised signal Hankel matrix is reconstructed by the singular values of signal subspace, and the denoised signal is finally extracted from this matrix. Experiment is performed to verify the feasibility of the proposed approach, and the results indicate that the proposed approach can reduce the Gaussian noise of PECT signal more effectively compared with other existing approaches.

Quantitative non-destructive evaluation of wall thinning defect in double-layer pipe of nuclear power plants using pulsed ECT method

In this paper, an inversion algorithm for three-dimensional profile reconstruction of wall thinning defect in a double-layer region of a typical coolant pipe of nuclear power plants from pulsed eddy current testing (PECT) signals has been proposed and experimentally validated, based on a fast simulator of PECT signals and a deterministic optimization strategy. First, the fast simulator developed by authors for PECT signal prediction based on a Fourier-series scheme in addition with interpolation and database approaches is briefly described as a base for the inversion of PECT signals. Then, the formula of the conjugate graduate inversion algorithm for sizing three-dimensional wall thinning from PECT signals is deduced in detail based on that for crack like reconstruction using single frequency ECT signals. The three-dimensional local wall thinning is modeled as a group of planar defects with different length and depth which are reconstructed from two-dimensional scanning PECT signals through inverse analysis. Through conducting PECT experiment for double-layer coolant tube test-piece and reconstructing wall thinning profile from the measured signals, the efficiency and the robustness of the proposed inversion algorithm are demonstrated.

An improved ferromagnetic material pulsed eddy current testing signal processing method based on numerical cumulative integration

A pulsed eddy current (PEC) testing signal processing method for ferromagnetic material is presented. Without any filtering process that may produce signal distortion, the numerical cumulative integration of the noisy PEC signal is calculated and subsequently fitted to the theoretical model. The model parameter estimation is later applied to quantify the specimen thickness and reconstruct a denoised PEC signal. A comparison is made between the presented method and the former direct-fitting method. The results demonstrate that the presented method improves the testing performance in terms of the detectable thickness range, the probe lift-off distance, and the measuring time.

A fast forward model of pulsed eddy current inspection of multilayered tubular structures

Multilayered Tubular Structures (MTS) such as pipeline are widely employed in such industrial fields as nuclear energy, petroleum, etc. It is vital to non-destructively evaluate MTS periodically before catastrophic accidents take place. As one of the advanced Electromagnetic Nondestructive Evaluation (ENDE) techniques, Pulsed Eddy Current Testing (PECT) has been found advantageous over the other ENDE methods regarding evaluation of integrity of MTS. In this paper, a fast forward model of PECT inspection of MTS is proposed based on the analytical modeling, namely the Extended Truncated Region Eigenfunction Expansion (ETREE). The closed-form expressions of PECT signals of 3D magnetic field and coil Electromotive Force (EMF) have been formulated. The proposed model has been verified by FiniteElement Modeling (FEM) and experiments. The advantages of the model in terms of high computational speed and accuracy have been identified.

Reconstruction of stress corrosion cracks using signals of pulsed eddy current testing

A scheme to apply signals of pulsed eddy current testing (PECT) to reconstruct a deep stress corrosion crack (SCC) is proposed on the basis of a multi-layer and multi-frequency reconstruction strategy. First, a numerical method is introduced to extract conventional eddy current testing (ECT) signals of different frequencies from the PECT responses at different scanning points, which are necessary for multi-frequency ECT inversion. Second, the conventional fast forward solver for ECT signal simulation is upgraded to calculate the single-frequency pickup signal of a magnetic field by introducing a strategy that employs a tiny search coil. Using the multiple-frequency ECT signals and the upgraded fast signal simulator, we reconstructed the shape profiles and conductivity of an SCC at different depths layer-by-layer with a hybrid inversion scheme of the conjugate gradient and particle swarm optimisation. Several modelled SCCs of rectangular or stepwise shape in an SUS304 plate are reconstructed from simulated PECT signals with artificial noise. The reconstruction results show better precision in crack depth than the conventional ECT inversion method, which demonstrates the validity and efficiency of the proposed PECT inversion scheme.

Sizing of Wall Thinning Defects Using Pulsed Eddy Current <newline/>Testing Signals Based on a Hybrid Inverse Analysis Method

Quantitative non-destructive evaluation, especially sizing of piping wall thinning in nuclear power plants is still a difficult and urgent issue. In this paper, an inversion approach for PECT (pulsed eddy current testing) signals is developed based on ANN (artificial neural network) method at first for profile reconstruction of wall thinning, the sizing result of NN is then utilized as the initial value of the CG (conjugate gradient) inversion scheme to overcome the shortages of both the NN (accuracy problem) and CG (local minimum problem) methods. Several reconstruction examples using the proposed hybrid strategy indicate that the combination of NN and CG methods is rather effective for wall thinning reconstruction from PECT signals in view of both the robustness and sizing accuracy.

An inversion scheme for sizing of wall thinning defects from pulsed eddy current testing signals

An inversion scheme for sizing of wall thinning defects from pulsed eddy current testing signals

Assessment of wall thinning in insulated ferromagnetic pipes using the time-to-peak of differential pulsed eddy-current testing signals

Pulsed eddy current testing (PECT) is a powerful candidate for the detection of wall thinning of insulated ferromagnetic pipes in petrochemical and power generation plants. The main purpose of this study is to find an efficient and easy-to-use signal feature for the assessment of wall thinning. Analytical modeling for a PECT probe over the insulated piping system is performed and its result is verified by experimental test. Two time-related features, the peak value and the time-to-peak, are found in the differential signal obtained by subtracting the test signal from the reference signal. The time-to-peak is superior to the peak value due to its linear variation with wall thickness. Influences of various conditions in practical testing on the PECT signal are investigated. Results show that the time-to-peak is independent of the insulation thickness and the probe lift-off. Robustness of time-to-peak to probe configuration is also validated by employing three probes of different dimensions and structures. To determine the linear range of time-to-peak with amount of wall thinning, differential signals based on different reference thicknesses are examined. Results show that the time-to-peak only keeps linear for the relative wall thinning less than 60%, but still can be effectively used for calibration purpose in periodical in-service inspection of insulated pipeline.

Development of a very fast simulator for pulsed eddy current testing signals of local wall thinning

A very fast numerical solver is developed for simulation of pulsed eddy current testing (PECT) signals caused by volumetric defects by introducing a database-type fast eddy current testing (ECT) simulation scheme for single frequency problems to the Fourier-series and interpolation-based PECT signal simulation code. First, the PECT signal simulation method based on a Fourier series scheme and an interpolation approach is briefly described. Then, a database-type fast numerical solver for single frequency ECT problems is introduced to the Fourier-series-based PECT simulation to enhance simulation efficiency. To cope with fast PECT signal simulation of 3D local wall thinning defects, a 2D shifting symmetry scheme is proposed for inspection targets of both plate and straight pipe geometries to reduce the computational burden required to establish databases of the unflawed field, which are necessary for fast ECT signal simulation. Using these strategies, a very fast numerical solver is developed for simulation of PECT signals of local wall thinning, and its validity is verified by comparing its numerical results with measured PECT signals and with those simulated using a conventional numerical code.