Fatigue Crack Monitoring Research Articles

Application of Bayesian estimation to structural health monitoring of fatigue cracks in welded steel pipe

Abstract Vibration induced fatigue is a well-known problem in oil and gas piping systems. However the use of vibration data to detect damage is not an easy task without a priori knowledge of the undamaged condition. In this paper, level 1: Detection and level 2: Localization of damage from structural health monitoring strategies is adapted for damage identification. An experimental trial of acoustic emission monitoring was run to monitor fatigue damage during a full-scale resonance fatigue test of a girth-welded steel pipe from healthy condition until failure. The pipe was excited into the first mode of vibration using a resonance fatigue testing machine in order to determine the high-cycle fatigue strength of the weld. The information provided by acoustic emission monitoring is useful in evaluating the condition of the pipe during the test and the occurrence of cracking before failure. However, the acoustic emission signals are embedded in noise. To overcome this problem, the signals from different combinations of sensors were recursively cross-correlated, which provides for the derivation of a new effective coefficient (EC) parameter for Bayesian estimation. This parameter is useful for evaluating uncertainty arising from the signals that contribute to source localization errors. The estimation finds the most probable parameters corresponding to cracks using prior knowledge derived from standard pencil lead break tests. The proposed method demonstrates a strong relationship between the acoustic emission energy and the estimated coefficients. A high correlation between signals was found to be associated with cracking, and a low correlation between signals was found to be associated with random signals or noise. The method will be useful for monitoring the condition of piping to manage the risk of vibration induced fatigue failure.

Continuous fatigue crack length estimation for aluminum 6061-T6 plates with a notch

Abstract This paper proposes an estimation technique of fatigue crack length for an aluminum 6061-T6 plate structure with a notch using a stress-life (S-N) curve. First, an S-N curve is obtained from multiple specimen tests by applying constant-amplitude cyclic loading with different stress levels. Next, the parameters of the Paris’ law are estimated from the S-N curve by solving an optimization problem. Then, based on the Paris’ law, the current crack length is continuously estimated by counting the elapsed fatigue loading cycles. The novelty of this study lies on the followings: (1) estimation of fatigue crack length using only the S-N curve for online fatigue crack monitoring, (2) estimation and experimental validation of the Paris’ law parameters estimated from the S-N curve, and (3) experimental validation of the crack length estimation technique for notched aluminum 6061-T6 plates with various thicknesses.

Thin-Film Sensor for Fatigue Crack Sensing and Monitoring in Steel Bridges under Varying Crack Propagation Rates and Random Traffic Loads

AbstractFatigue cracks are critical structural concerns for steel highway bridges, and fatigue initiation and propagation activity continues undetected between physical bridge inspections. Monitori...

A conductive grating sensor for online quantitative monitoring of fatigue crack.

Online quantitative monitoring of crack damage due to fatigue is a critical challenge for structural health monitoring systems assessing structural safety. To achieve online quantitative monitoring of fatigue crack, a novel conductive grating sensor based on the principle of electrical potential difference is proposed. The sensor consists of equidistant grating channels to monitor the fatigue crack length and conductive bars to provide the circuit path. An online crack monitoring system is established to verify the sensor's capability. The experimental results prove that the sensor is suitable for online quantitative monitoring of fatigue crack. A finite element model for the sensor is also developed to optimize the sensitivity of crack monitoring, which is defined by the rate of sensor resistance change caused by the break of the first grating channel. Analysis of the model shows that the sensor sensitivity can be enhanced by reducing the number of grating channels and increasing their resistance and reducing the resistance of the conductive bar.

Utilising hydraulic transient excitation for fatigue crack monitoring of a cast iron pipeline using optical distributed sensing

Utilising hydraulic transient excitation for fatigue crack monitoring of a cast iron pipeline using optical distributed sensing

Thermography Applied to the Study of Fatigue Crack Propagation in Polycarbonate

Thermography was used to study the propagation of fatigue cracks during cyclic loading of pre-cracked SAE keyhole polycarbonate specimens. A micro-bolometer infrared camera (FLIR A655sc) and a commercially available software program (DeltaTherm2) were employed. The stress intensity factors were determined using a hybrid thermoelastic stress analysis (TSA) technique. The crack growth rate was determined via thermography using two different approaches. The first approach used the output of the crack-tip position from the developed TSA algorithm and the number of cycles between data sets. The second approach used temperature measurement as a new way to determine da/dN (crack growth rate) directly. As a result, da/dN vs ΔK (stress intensity factor range) graphs were plotted and fitted using Paris’ law. A comparison between the resultant da/dN vs ΔK curves and results found in the literature, as well as curves from the finite element method (FEM) simulations showed good agreement. The conclusion was that thermography is a very powerful tool that can detect, measure and monitor fatigue cracks in polycarbonate.

Low Computation Acoustic Emissions Structural Health Monitoring Through Analog Signal Pre-Processing

In this research an innovative acoustic emission sensing system has been developed intended for structural fatigue crack monitoring. The innovation lies in analog pre-processing of the detected acoustic emissions for signal enveloping, thus relying on cheap high bandwidth components. The technique is based on heterodyning the amplified acoustic emission signal with a carrier signal of a chosen frequency. Next, the signal is filtered and the signal envelope is obtained by phase shifting the signal by pi/2 creating an analytic signal that is digitally sampled. This results in need of low sampling rate digital acquisition equipment giving relatively small amounts of data to be processed and stored, considerably reducing the system cost. This is particularly suitable for applications involving large or complex structures to be monitored, where a multitude of sensors are needed. The system was built and tested on aluminum test coupons during tension-tension fatigue. The envelope signal is filtered for background noise through thresholding based on statistical knowledge of the noise distribution. The accumulation of acoustic activity shows promising results with an early period of high acoustic activity during settling of the material that asymptotically converges at a certain level. At the next stage, there is no activity until a certain point is reached where a sudden ramp up of AE is detected close to the end of the experiment. Throughextensometer measurement, the change in coupon length at this point in time strongly indicates that the ramp up of AE activity is due to crack initiation and propagation.

Fatigue crack monitoring in train track steel structures using plastic optical fiber sensor

Plastic optical fiber (POF) sensors have shown excellent potential for damage detection and structural health monitoring in a variety of engineering structures. This paper discusses the feasibility of using POF sensors in conjunction with a signal-processing algorithm capable of detecting and monitoring fatigue-induced cracks in train track steel structures in real time. The POF sensor, which was modified from an existing design to increase the signal sensitivity, allows for accurate detection of a fatigue crack developed in a specimen, and was found to compare well to the reference acoustic emission (AE) sensors and crack opening displacement (COD) gauge attached to the specimen. The crack-detection technique, which relies on capturing the intensity variation of the POF sensor, was not susceptible to any signal fluctuations commonly associated with intensity-based optical fiber sensors. The results show that the technique has potential for use in detecting the initiation and propagation of specific segments of a structure vulnerable to cracking due to external cyclic loading, e.g. at welded joints in train tracks under train loads or offshore structures subject to wave loads. The POF sensor system is composed of inexpensive parts (LED light source, photodetectors, and data acquisition units) and can easily be installed to the host structure. To validate the proposed damage-detection technique, the instrumented specimens are subjected to cyclic loading in order to induce stable crack propagation in the specimen. A COD gauge and AE were used for the purpose of calibration and comparison. The results show remarkable resemblance in terms of crack initiation and propagation identification exhibited by all three types of sensors, highlighting the potential of the proposed sensor for crack initiation detection and subsequent monitoring of crack propagation.

Analysis of load effect on nonlinear vibro-acoustic modulation used in on-line monitoring of fatigue cracks

Vibro-acoustic modulation is a convenient and effective nondestructive detecting approach. However, when using this technology to monitor structures under in-service conditions, conventional damage indices will not be able to effectively track the damage propagation due to the detrimental influence induced by service load. In this paper, an analysis model was approached based on plasticity induced crack closure stress analysis. This model reveals the linear dependence between the crack size and the changing rate of contact area with respect to external applied load. A load insensitive damage index (LIDI) was then developed aiming to track the crack growth under varying applied load. Results of on-line experiments confirm the load effect on the conventional damage index and prove that the LIDI is able to track the crack propagation under varying applied load. Furthermore, experimental results also demonstrate that the LIDI shows a higher sensitivity of small crack and also other improved tracking performance compared with the conventional one. We believe the LIDI holds a great potential in early damage inspection of in-service structures.

Fatigue crack monitoring of aerospace structure based on lamb waves and binary tree support vector machines

To overcome the difficulty in identifying the fatigue crack in key parts of aerospace structure, a kind of methods aimed to monitor the crack length based on matching pursuit (MP) method and binary tree support vector machines (BT-SVM) classification algorithm was developed. In this method, Lamb wave signals were decomposed into a linear combination of several Chirplet atoms by MP method, and then the matching parameters were extracted as feature vectors for training and testing in BT-SVM classification algorithm. At the same time, a lug joint model was created with a certain ratio and the effect of crack extension on Lamb wave signals propagation was studied. At last, fatigue loading experiments were carried out in lug joints and tail reinforced frames of aircraft. The results showed that this method can monitor the length of fatigue crack effectively, which presents a new approach for monitoring the fatigue crack.

Bayesian approach to breathing crack detection in beam structures

In this paper, a Bayesian approach is developed to conduct uncertainty quantification on a single breathing crack in a beam structure using nonlinear forced responses. The proposed methodology not only determines the breathing crack characteristics but also quantifies associated uncertainties of the inferred values. Such information is important for fatigue crack monitoring and remaining life prediction in cracked beam structures. First, a single degree of freedom model is developed to characterize the nonlinear behavior of the cracked beam. The Modified Homotopy Perturbation Method (MHPM) is applied to determine analytical approximate solutions. Then, a Bayesian inference approach is proposed by applying Markov chain Monte Carlo (MCMC) technique, in which the Random Walk Metropolis algorithm is employed. The objective is to estimate crack size or location from the nonlinear vibration responses, in which noise is added to represent actual measurement data. Finally, the proposed probabilistic damage detection approach is successfully demonstrated and the breathing crack status is quantified with associated uncertainties. This leads to a new way of detecting a single breathing crack in beam structures.

A large-area strain sensing technology for monitoring fatigue cracks in steel bridges

This paper presents a novel large-area strain sensing technology for monitoring fatigue cracks in steel bridges. The technology is based on a soft elastomeric capacitor (SEC), which serves as a flexible and large-area strain gauge. Previous experiments have verified the SEC’s capability to monitor low-cycle fatigue cracks experiencing large plastic deformation and large crack opening. Here an investigation into further extending the SEC’s capability for long-term monitoring of fatigue cracks in steel bridges subject to traffic loading, which experience smaller crack openings. It is proposed that the peak-to-peak amplitude (pk–pk amplitude) of the sensor’s capacitance measurement as the indicator of crack growth to achieve robustness against capacitance drift during long-term monitoring. Then a robust crack monitoring algorithm is developed to reliably identify the level of pk–pk amplitudes through frequency analysis, from which a crack growth index (CGI) is obtained for monitoring fatigue crack growth under various loading conditions. To generate representative fatigue cracks in a laboratory, loading protocols were designed based on constant ranges of stress intensity to limit plastic deformations at the crack tip. A series of small-scale fatigue tests were performed under the designed loading protocols with various stress intensity ratios. Test results under the realistic fatigue crack conditions demonstrated the proposed crack monitoring algorithm can generate robust CGIs which are positively correlated with crack lengths and independent from loading conditions.

Real-time structural health monitoring of fatigue crack on aluminum beam using an impedance-based portable device

A common problem in aircraft maintenance is the development and detection of fatigue cracks in structural components with localized high-stress concentration. Traditional non-destructive testing methods are effective in inspecting cracks through an offline manner; however, real-time implementation of these methods is not common. In this contribution, a procedure to monitor propagating cracks in aluminum beams undergoing dynamic cyclic loading is presented. High-frequency impedance measurements were performed from a lead zirconate titanate patch bonded to an aluminum beam while the beam experienced cyclic loading at a prescribed amplitude. Fatigue loading at prescribed stress levels was achieved by exciting the beam with an electromechanical shaker. A control scheme was used to keep the drive frequency and drive acceleration under resonance in the first vibration mode. During the tests, impedance measurements were acquired using a low-cost portable impedance device. Damage indices were correlated with the specific number of stress cycles until changes in the natural frequency caused by propagating fatigue cracks reached a critical limit. Finally, a statistical method is proposed to obtain the damage threshold. Based on S/N curves, the determined threshold was compared to the damage indices so that incipient damage could be detected before a visible crack could be seen.

Mixture Proposal Particle Filtering for Guided Wave Based Fatigue Crack Propagation Prognosis

For on-line prognosis of fatigue crack propagation with guided wave based structural health morning, uncertainties always exist in the process of fatigue crack propagation and crack monitoring. The particle filtering is demonstrated as a powerful tool to deal with these uncertainties. However, most literatures adopt the standard sequential importance sampling resampling algorithm, which uses the transition probability density function as the importance density and neglects current measurement information. Satisfactory results may be unable to be achieved if particles are drawn from an inferior transition probability density. This paper proposes a kind of mixture proposal distribution as the importance density that linearly combines the measurement probability density with the transition probability density. In addition, the damage index of the on-line monitored guided wave signal is adopted for quantifying the crack length. Each time a new damage index is available, the mixture proposal particle filtering is adopted to evaluate the posterior probability density estimate of the crack length, as well as the crack propagation prognosis. Simulated crack propagation data is used for validation, and the result shows that the mixture proposal particle filtering outperforms the standard particle filtering algorithm.

Fatigue crack monitoring using plastic optical fibre sensor

Optical fibre sensors have shown their excellent potential for structural monitoring in a variety of engineering structures. This paper discussed the feasibility of real-time monitoring of fatigue crack using an extrinsic plastic optical fibre sensor. This intensity-based plastic optical fibre (POF) sensor was fabricated by connecting two cleaved optical fibres into a capillary tube and inserting some in-filling into the tube to improve the sensitivity of the sensor. The aim of this sensor is to detect crack initiation and subsequently monitor the propagation of crack under fatigue load, relying only on the plastic fibre sensor signal and a simple signal processing method. Performance of the fabricated POF sensor was evaluated by comparing with another technique, namely, acoustic emission (AE) method. The results show remarkable resemblance in terms of crack initiation and propagation identification exhibited by both of the sensors, highlighting the potential of the proposed method for crack initiation detection and subsequent monitoring of crack propagation. The simplicity of the technique renders it an attractive structure health monitoring tool for fatigue crack monitoring.

Real-Time Fatigue Crack Monitoring Based on Fiber Bragg Grating Sensors

In this paper, an real-time fatigue crack monitoring method is investigated based on the fiber Bragg grating (FBG) sensors. Three FBGs are applied along the direction of crack propagation to monitor a fatigue crack with size of 10mm. Experimental results demonstrate that the reflection spectrum changes observably with the propagation of crack when the crack is within 3-4mm long of each FBG. Characteristic parameters are constructed to detect the variation degree of reflection spectrum. A crack quantification model based on characteristic parameters is proposed to describe the correlation between the crack size and damage sensitive features extracted form FBG measurements. It makes possible to monitor the propagation of crack quantitatively within 3-4 mm long for each FBG. Finally, a certain size crack can be monitored using several FBGs.

Measuring fatigue crack growth and closure in Polyether Ether Ketone (PEEK)

Throughout industry, polymers are being used to replace parts that were previously fabricated from metals. Thus, the fatigue resistance of polymers needs to be well understood. Polyether Ether Ketone (PEEK) is often chosen for its good mechanical properties and high glass transition temperature. In this paper, the experiments and methods to study fatigue crack growth and subsequent crack closure are presented. Compact specimens made of PEEK were used to experimentally monitor fatigue crack growth at R=0.1 and R=0.7. Some unique aspects of the experiments include the use of compression pre-cracking and a back-face strain gauge to measure crack length. While measuring crack closure was the primary focus of the experiments, crack growth rate versus stress intensity factor range results were also generated. Using remote methods of measuring crack closure show that PEEK experiences a small level of crack closure, though the effects are minimal. However, indirect methods of measuring crack closure show that the phenomenon is more prevalent in the material than what was measured. Further research for continuing the study of fatigue crack closure in PEEK are discussed.

Numerical simulation and experimental validation of a large-area capacitive strain sensor for fatigue crack monitoring

A large-area electronics in the form of a soft elastomeric capacitor (SEC) has shown great promise as a strain sensor for fatigue crack monitoring in steel structures. The SEC sensors are inexpensive, easy to fabricate, highly stretchable, and mechanically robust. It is a highly scalable technology, capable of monitoring deformations on mesoscale systems. Preliminary experiments verified the SEC sensor’s capability in detecting, localizing, and monitoring crack growth in a compact specimen. Here, a numerical simulation method is proposed to simulate accurately the sensor’s performance under fatigue cracks. Such a method would provide a direct link between the SEC’s signal and fatigue crack geometry, extending the SEC’s capability to dense network applications on mesoscale structural components. The proposed numerical procedure consists of two parts: (1) a finite element (FE) analysis for the target structure to simulate crack growth based on an element deletion method; (2) an algorithm to compute the sensor’s capacitance response using the FE analysis results. The proposed simulation method is validated based on test data from a compact specimen. Results from the numerical simulation show good agreement with the SEC’s response from the laboratory tests as a function of the crack size. Using these findings, a parametric study is performed to investigate how the SEC would perform under different geometries. Results from the parametric study can be used to optimize the design of a dense sensor network of SECs for fatigue crack detection and localization.

Monitoring fatigue cracks of a metal structure using an eddy current sensor

The present paper investigates monitoring of fatigue cracks of a metal structure using an eddy current micro sensor. Fatigue cracks tend to occur at bolt-jointed structures on an aircraft. In order to detect the damage quantitatively, a kind of change-prone micro eddy current sensor is designed and fabricated with flexible printed circuit board (FPCB) technology. A forward semi-analytical model is built by extracting a material’s conductivity as the damage feature parameter, and characteristics analysis is conducted based on the model. The research focuses on setting up and utilizing the eddy current fields to analyze interaction of adjoining coils when the damage occurs, and investigating optimization on the working parameters of the sensor. In the experimental section, several common connection structures are applied to explore the sensor’s monitoring ability both in air and in a corrosive environment. The result shows that the optimal working frequency is about 1 MHz. The eddy current micro sensor is capable of monitoring the crack growth with an accuracy of 1 mm, the average error being 4.6 % compared to fracture analysis. The sensor keeps high resolution of damage in aqueous corrosion. Due to the fretting fatigue, wear appears on the polyimide foil, leading to the decreases of the monitoring signal.

Evaluation of fatigue crack network growth in cast iron for different biaxial loading paths via full-field measurements

This paper proposes a new method for monitoring fatigue crack network initiation and growth based on Digital Image Correlation (DIC), and reports on in-plane biaxial fatigue experiments with two different loading paths and two amplitudes on nodular graphite cast iron. The central thinned part of cross-shaped samples is observed on two scales. Regularized Digital Image Correlation (DIC) is used to measure displacement fields and to reveal DIC residuals, i.e., image differences that cannot be accounted for with registration. The detailed strain histories are compared at different scales and for the different loading regimes. DIC and mechanical residuals, i.e., local deviations of the displacement field from being the solution to an elastic problem, enable for the crack network detection and quantification with respect to the number of cycles. The relative damage severity and fatigue lifetime of different loading paths experiencing identical load magnitudes are analyzed for the tested material.