Robust Non-destructive Evaluation Research Articles

Dynamic Defect Detection in Fast, Robust NDE Methods by Transfer Learning Based Optimally Binned Hypothesis Tests

ABSTRACT The structural integrity of safety-critical infrastructures diminishes over time, necessitating periodic inspections. The gathering of low-noise, precise nondestructive evaluation (NDE) readings for defect detection can be both time-consuming and costly. In practical applications, NDE data can be highly noisy due to factors like liftoff/stand-off distances, probe drift, scanning speed, and variations in data acquisition rates. Given the broad spectrum of potential uncertainties that can contribute to noise accumulation, characterizing the noise in such NDE data presents significant difficulties. As a result, the extraction of defect signals through deconvolution becomes challenging. This article introduces an NDE-based dynamic defect tracking framework designed for robust conditions. It incorporates regular periodic monitoring of the material under test (MUT) using a cost-effective, easy-to-implement magnetic flux leakage (MFL) probe. The framework proposes a dynamic setup that includes occasional, highly accurate, expensive MFL scans among regular, low-cost, noisy MFL scans. The objective is to detect and precisely track defect formation in metallic pipes over time, thereby enabling the efficient alarming and localization of defects before they reach detrimental sizes. A key characteristic of the proposed dynamic monitoring method is its noise type agnosticism in the low-cost MFL scans. It remains effective even when the underlying noise-producing uncertainties fluctuate significantly over time, a feature achieved through the use of transfer learning. The method transfers accurate information regarding the locations and sizes of existing corrosions from the detailed MFL scan to the subsequent noisy MFL inspections. With this supportive information from defective scan-points, it estimates the degradation of the signal-to-noise ratio (SNR) in the noisy MFL scans. The framework suggests binned hypothesis tests in noisy MFL scans, setting the level of aggregation based on the SNR estimate in the scan data. This proposed binned hypothesis test optimizes defect coverage and maintains control over the false discovery rate (FDR). Experimental MFL data, gathered under a wide range of uncertainties and defect types, illustrates the highly heterogeneous nature of noise distributions in low-cost MFL scans and the significant variance in SNR deteriorations. The necessity of using binned hypothesis tests in these low-cost MFL data is highlighted by the poor performance of pointwise multiple hypothesis tests. Finally, the high effectiveness of the proposed transfer learning-based binned hypothesis testing method is demonstrated. Although the method was applied to MFL data in this instance, it is also suitable for other NDE applications. The applicability of the algorithms has also been confirmed through validation on experimentally generated Eddy Current (EC) data.

Monitoring of crack healing in geopolymer concrete using a nonlinear ultrasound approach in phase-space domain

Ultrasonic testing is a robust non-destructive evaluation method frequently employed in the health monitoring of concrete structures. Cracking in concrete can be one of the most critical problems, and its healing is important for structural safety. The current study proposes the evaluation of crack healing in geopolymer concrete (GPC) using different linear and nonlinear ultrasonic techniques. In this regard, a notched GPC beam was constructed at the laboratory and geopolymer grout was applied as a repair material. Ultrasonic pulse velocity (UPV) and signal waveform tests were performed at several stages before and after grouting the notch. Nonlinear wave signals were processed in the phase-space domain for qualitative health monitoring of GPC. Furthermore, feature extraction was applied to phase-plane attractors using fractal dimension for quantitative assessment. The sideband peak count-index (SPC-I) method was also utilized to assess the ultrasound waves. The results indicate that the phase-space analysis of ultrasound can successfully represent the healing progress inside the GPC beam. At the same time, the fractal dimension can be used as a healing index. Ultrasound signal attenuation demonstrated high sensitivity to crack healing. The SPC-I technique exhibited an inconsistent trend at the early stage of healing. However, it provided a clear indication of repair at the advanced stage. Although the linear UPV method was found sensitive to the grouting at the initial stages, it demonstrated insufficient capability to monitor the healing process fully. Therefore, the phase-space-based ultrasonic method and the attenuation parameter could be employed as reliable techniques for the robust monitoring of progressive healing in concrete.

Robust defect detection under uncertainties using spatially adaptive capacitive imaging

We develop a high-Q capacitive sensing based robust non-destructive evaluation (NDE) methodology that can be widely used in varied NDE applications. We show that the proposed method can detect defects in a host of robust regimes where uncertainties such as lift-off, probe tilt, fluctuations in sampling rates, and step sizes are inherent in the data collection process. We explicitly characterize the corruption in the capacitive sensing data due to various lift-off based uncertainties. We use a Bayesian decision theoretic approach to rigorously understand the impact of these corruptions on defect identification efficacy. Using an optimally tuned weighted classification loss, we prove that it is theoretically feasible to accurately detect defect location and sizes from capacitive sensing signals collected under the aforementioned uncertainties. The Bayesian decision theoretic study needs prior information for accurate detection that is not available in real NDE inspections. So, we develop a solely data driven algorithm that analyzes the capacitive sensing signals without any prior knowledge of defect or uncertainty types. The developed algorithm is non-parametric and uses spatially adaptive denoising to weed out uncertainty induced noises. By leveraging the spatial association in the capacitive sensing signals, our algorithm greatly improves on popular non-spatial approaches. Compared to popular thresholding methods and low-rank based denoising approaches, we demonstrate superior performance of the proposed method in terms of coverage and false positive metrics for defect identification. Using spatially adaptive denoising, we design a robust capacitive sensing method that can detect defects with high precision under various uncertainty regimes.

Nondestructive evaluation of resistance spot-welded Al-steel joints

Abstract

Quality Assessment of Friction Stir Welded Joints – Using the Fundamental Anti-Symmetric Lamb Wave Mode

Having a robust non-destructive evaluation (NDE) technique for friction stir welded (FSWed) joints is of interest to the processing community. Such a technique has to be sensitive to the different types and shapes of internal weld defects and has to be applicable for both similar and dissimilar material FSW joints. Investigated was the ability of ultrasonic guided waves to detect and assess the quality of FSW joints. The fundamental anti-symmetric (A0) mode was selected to detect the flaws in FSW joints. Guided waves were excited (using PZT wafers) and received (using a laser Doppler vibrometer, LDV). Implemented was the frequency-wavenumber filtering technique to separate forward propagating wave from any back propagating reflected wave due to the welded joint. Identified was the reflection of the A0 mode caused by the presence of the interface and/or defects within the joint. The findings indicate little sensitivity to the presence of material interface suggesting this technique to have a promising potential among guided-wave-based techniques in the qualitative and quantitative assessment of FSW joints.

Detection and assessment of pitting corrosion in rebars using scattering of ultrasonic guided waves

The present study aims at developing a robust non-destructive evaluation technique by using the scattering of ultrasonic guided waves for early detection and assessment of localised damage in bare rebars in the form of corrosion pits. The guided wave interaction with the pitting corrosion is numerically simulated by using finite element analysis, and the results are validated with experiments. The received signal contains several wave packets, which could be identified as various scattered modes that are generated from the edges of the corroded region. It is found that when the damage is placed at the centre of the rebar, several scattered wave modes superimpose to form two distinct wave packets in the received signal. A damage index is proposed by using these two distinct wave packets. It is shown that the proposed damage index method can be used successfully to monitor the axial extent and intensity of pitting corrosion in rebars. However, for the pitting corrosion that is situated at an arbitrary position, a different methodology, which utilises the differential time of arrival of various scattered wave modes, is proposed to identify the location of the pitting corrosion. Finally, a non-dimensional parameter, namely the scatter coefficient, is defined to analyse the energy contributions of various scatter modes. It is found that this scatter coefficient gradually increases with the mass loss and that the increment is pronounced when the corrosion pit is closer to the receiver location. The findings of the study can be used to identify the pitting corrosion in bare rebars.

A computational approach for robust nondestructive test design maximizing characterization capabilities for solids and structures subject to uncertainty

SummaryA robust approach to nondestructive test (NDT) design for material characterization and damage identification in solids and structures is presented and numerically evaluated. The generally applicable approach combines maximization of test sensitivity with minimization of test information redundancy, while simultaneously minimizing the effects of uncertain system parameters to determine optimal NDT parameters for robust nondestructive evaluation. In addition, to maintain reasonable computational expense while also allowing for general applicability, a stochastic collocation technique is presented for the quantification of uncertainty in the robust design metrics. The robust NDT design approach was tested through simulated case studies based on the characterization of localized variations in Young's modulus distributions in aluminum structural components utilizing steady‐state dynamic surface excitation and localized measurements of displacement and compared with a purely deterministic NDT design approach. The robust design approach is shown to produce substantially different NDT designs than the analogous deterministic strategy. More importantly, the robust NDT designs are shown to provide significant improvements in the ability to accurately nondestructively evaluate structural properties for the cases considered, when there is significant uncertainty in system parameters and/or aspects of the NDT implementation. Copyright © 2015 John Wiley & Sons, Ltd.

A Thickness Nondestructive Evaluation Strategy for Coating Located in Truncated Conical Steel Shells of Various Sizes Based on Eddy Current Testing Technology

The anti-corrosive coating in oil tanker is a crucial part. Its thickness needs a robust non-destructive evaluation (NDE) tool to estimate. To get satisfied results from eddy current signals based on a limited number of samples, the influence of test-piece sizes on evaluation precision of liftoff is analyzed by using a FEM–BEM hybrid program, and a NDE strategy based on sample database is proposed. Its applicability is verified by simulation results.

Stress measurements via photoluminescence piezospectroscopy on engine run thermal barrier coatings

Optical measurements of stress in a thermally grown oxide (TGO) layer that has formed between a metallic bond coat and thermal barrier coating (TBC) have been previously demonstrated and shown useful in understanding aspects of TBC failure. These measurements have promise for nondestructive evaluation of coated turbine components. However, engine-run turbine parts collect significant surface contamination consisting of calcium-magnesium-alumina-silicate (CMAS). The deposited CMAS both blocks optical measurement of stress in the TGO and produces false stress signals. A recently developed laser ablation technique has enabled contaminant removal with minimal TBC damage. The locally cleaned engine run parts can then have their TGO stresses optically determined. In the present paper, it is shown that sporadic signals from contamination are still present due to CMAS infiltration into the TBC and due to local regions that cannot be cleaned without damaging the TBC. Methods for separating the contamination signal from the TBC signals are presented as needed for successful use of optical stress measurement on engine run turbine components. The stress measurements censored for false signals are compared to other destructive approaches and the methods are shown to provide robust nondestructive evaluation of the coating stress.

Tile-wall bonding integrity inspection based on time-domain features of impact acoustics

The objective of this paper is to investigate the acoustics characteristics obtained from impact sounds excited by impact on tiled-walls and hence to develop a rapid, effective and robust non-destructive evaluation method for bonding integrity inspection. Most of the previous research on impact acoustics evaluation relies on the frequency-domain characteristics of the free-vibration component of impact sounds for detection and feature extraction. This method proves to be too sensitive to the surface roughness of the practical structures. In the present work, special attention is paid to the time-domain acoustics to present a solution tolerant of high surface non-uniformity. Based on the impact dynamics analysis, the relationship between the acceleration and ringing components of impact sounds and the corresponding bonding state is firstly clarified. Investigation performed on the pattern of the initial acceleration component in the time record reveals its effectiveness in indicating the existence of bonding defects in tile-walls with satisfactory immunity to surface roughness. By properly selecting the feature vectors extracted from the initial acceleration peaks of the impact sounds, a multilayer back-propagation artificial neural network (ANN) classifier is developed for automatic health monitoring of the tile-walls. The inspection results obtained experimentally on sample slabs are presented and discussed, confirming the validity of the proposed method.

Changes in electrochemical impedance with microstructural development in TBCs

Thermal barrier coatings (TBCs) are improving the performance and efficiency of advanced gas turbine engines by allowing higher inlet temperature and insulation of critical hot-section components. Monitoring the integrity of TBCs prior to failure is critical to the overall performance of gas turbine en gines and requires a robust nondestructive evaluation (NDE) technique. In this paper, changes in electrochemical impedance with microstructural degradation of critical constituents in TBCs are summarized for the development of electrochemical impedance spectroscopy as an NDE technique for TBCs.

Monitoring damage evolution in thermal barrier coatings with thermal wave imaging

Thermal wave imaging (TWI) is a promising non-destructive evaluation (NDE) technique with the ability to detect integrity and thickness of thermal barrier coatings (TBCs) extensively used for advanced gas turbine engines. Development of a robust NDE technique is essential for quality control, life assessment, and health monitoring of TBCs for applications, maintenance, and prevention of catastrophic failure. In this study, TWI was employed as an NDE technique to examine as-coated TBCs with varying thicknesses, and thermally cycled TBCs for initiation and progression of subcritical–subsurface damage as a function of thermal cycling. TBC specimens examined consisted of air plasma sprayed ZrO 2–7 wt.% Y 2O 3, NiCoCrAlY bond coats and Haynes 230 superalloy. Thermal cycling was carried out in air with 30-min heat-up, 10-h dwell at 1150 °C, 30-min air-quench and 1-h hold at room temperature. During thermal cycling, TBC specimens were evaluated non-destructively with TWI at room temperature every 10 to 20 thermal cycles, and selected specimens were removed from thermal cycling for microstructural analysis by scanning electron microscopy (SEM). TWI analysis was correlated to the microstructural characteristics and damage progression of TBCs based on phenomenological expressions of thermal diffusion. Higher thermal response amplitude associated with disrupted heat transfer was observed where localized spallation at or near the YSZ/TGO interface occurred.