Adam C. Cobb

A comparison of feature-based classifiers for ultrasonic structural health monitoring

Diffuse ultrasonic signals received from ultrasonic sensors which are permanently mounted near, on or in critical structures of complex geometry are very difficult to interpret because of multiple modes and reflections constructively and destructively interfering. Both changing environmental and structural conditions affect the ultrasonic wave field, and the resulting changes in the received signals are similar and of the same magnitude. This paper describes a differential feature-based classifier approach to address the problem of determining if a structural change has actually occurred. Classifiers utilizing time and frequency domain features are compared to classifiers based upon time-frequency representations. Experimental data are shown from a metallic specimen subjected to both environmental changes and the introduction of artificial damage. Results show that both types of classifiers are successful in discriminating between environmental and structural changes. Furthermore, classifiers developed for one particular structure were successfully applied to a second one that was created by modifying the first structure. Best results were obtained using a classifier based upon features calculated from time-frequency regions of the spectrogram.

Self‐Calibrating Ultrasonic Methods for In‐Situ Monitoring of Fatigue Crack Progression

Ultrasonic sensors permanently affixed to aluminum coupons are used to monitor progression of damage during fatigue testing with the long term goal of structural health monitoring for diagnostics and prognostics. Necessary for success are proper design of the ultrasonic testing methods, robust transducer mounting techniques, and real‐time signal processing for determining the state of the structure. It is also highly desirable for the overall system to be self‐calibrating with built‐in diagnostics in order to detect and compensate for sensor degradation or failure. Self‐calibrating ultrasonic techniques are applied for monitoring of cracks initiating and propagating from the inaccessible inner diameters of rivet holes where the transducers are mounted on the accessible specimen surface. Angle beam ultrasonic methods are utilized that are suitable for detecting small defects in critical local regions of high stress. Results are presented for aluminum coupons subjected to low cycle fatigue and demonstrate u...

ULTRASONIC STRUCTURAL HEALTH MONITORING: A PROBABILITY OF DETECTION CASE STUDY

Ultrasonic interrogation of metal alloys has been demonstrated to be effective for monitoring fatigue‐induced damage in a structural health monitoring (SHM) framework. Before such a method can be implemented, the probability of detection (POD) as a function of crack size must be quantified. POD curves are routinely generated for nondestructive evaluation (NDE) methods, typically by performing large numbers of measurements to capture the variability arising from variations in operators, probes, instruments and crack morphology. Such studies have not yet been carried out for many, if any, SHM methods, and thus identifying and quantifying relevant sources of variability have not generally been addressed. Considered here is monitoring of fastener holes for fatigue cracks, and POD curves are generated using essentially the same methods as are used for NDE but with differences in setting detection thresholds. Interpretation of the curves is discussed given that operator, sensor and instrumentation variability a...

Review of magnetostrictive transducers (MsT) utilizing reversed Wiedemann effect

Magnetostrictive transduction has been widely utilized in NDE applications, specifically for generation and reception of guided waves for long-range inspection of components such as pipes, vessels, and small tubes. Transverse-motion guided wave modes (e.g., torsional vibrations in pipes) are the most typical choice for long-range inspection applications because the wave motion is in the plane of the structure. Magnetostrictive-based sensors have been available for several years for these wave modes based on the Wiedemann effect. For these sensors, a permanent magnetic bias is applied that is perpendicular to the direction of the propagated guided wave. This bias field strains the material that the guided wave is generated in preferentially in the desired particle motion direction. A time-varying magnetic field oriented parallel to the direction of guided wave propagation is also induced in the material. This time-varying field is induced using an electric coil located near the material surface. The intera...

Ultrasonic Sensing and Life Prediction for the DARPA Structural Integrity Prognosis System

The overall objective of the DARPA Structural Integrity Prognosis System (SIPS) program is to develop technologies to advance material damage state condition assessment with limited or no dedicated maintenance action. As a part of the sensors thrust area, an in situ ultrasonic sensing method was developed and demonstrated to detect cracks initiating from fastener holes and provide an estimate of total crack area. Crack area estimates were combined with load history data, projected future loads, and life prediction models to determine a probability density function for time‐to‐failure. The ultrasonic method utilizes two shear wave angle beam transducers operating in through transmission mode which are mounted on either side of the hole. The transmitted wave travels through the area of expected cracking, and the presence of cracks around the fastener holes decreases the amount of acoustic energy that is received. Furthermore, as cracks open and close during the fatigue process, the received energy is modula...

A BAYESIAN APPROACH FOR ESTIMATING SIZES OF FATIGUE CRACKS NEAR FASTENER HOLES

Ultrasonic in situ monitoring of metal alloys has been successfully demonstrated for determining the presence and size of fatigue damage within a structure. Ultrasonic techniques, however, only provide an estimate of the state of the structure at that time and do not predict the remaining fatigue life. On the other hand, a statistical crack propagation approach, which models the expected remaining life based on an assumed fatigue process, specimen geometry and material properties, allows for the fatigue life to be estimated. To maintain the safety of the structure, this approach typically requires assuming a worst case initial flaw size. Presented here is a Bayesian estimation approach for incorporating both the measurement and modeling methodologies. An Extended Kalman Filter approximation is used to combine ultrasonic estimates of fatigue cracking with a crack propagation model. The measurement model is based upon recent work by the authors on a shear wave, angle‐beam method for monitoring fastener hole...

An integrated approach to local ultrasonic monitoring of fastener hole fatigue cracks

Ultrasonic nondestructive evaluation methods are routinely used to detect and size fatigue cracks near fastener holes in aircraft structures as a part of scheduled maintenance. In contrast, statistical crack propagation models provide an estimate of the expected fatigue life assuming a known crack size and future fatigue loadings. Here an integrated approach for in situ diagnosis and prognosis of fastener hole fatigue cracks is proposed and implemented that incorporates both ultrasonic monitoring and crack growth laws. The sensing method is an ultrasonic angle beam technique, and cracks are automatically detected from the ultrasonic response. An extended Kalman filter is applied to combine ultrasonically estimated crack sizes with a crack growth law, effectively using the time history of the ultrasonic results rather than only the most recent measurement. A natural extension of this method is fatigue life prognosis. Results from fatigue tests on 7075-T651 aluminium coupons show improved crack size estimates as compared to those obtained from ultrasonic measurements alone, and also demonstrate the capability of predicting the remaining life. This approach for fatigue crack detection, sizing and prognosis is an example of a general strategy for in situ monitoring of structural damage whereby improved results are achieved from the integration of noisy measurements with imperfect crack growth models.

SIMULTANEOUS ULTRASONIC MONITORING OF CRACK GROWTH AND DYNAMIC LOADS DURING A FULL SCALE FATIGUE TEST OF AN AIRCRAFT WING

An in situ ultrasonic angle beam method has been developed to detect and size fatigue cracks emanating from fastener holes. The method employs transducers mounted in a pitch‐catch configuration on each side of the hole, and the modulation of received signals with load is used to detect and size small fatigue cracks. The focus of this paper is the monitoring of a series of critical fastener holes during a full scale fatigue test of an aircraft wing. Essential to the success of this method is the ability to use the same ultrasonic signal as is used for crack monitoring to estimate the instantaneous load. The applied load causes a time shift in ultrasonic echo arrivals resulting from both dimensional changes and the acoustoelastic effect. Waveforms are recorded randomly during fatiguing, and the dynamically applied load associated with each waveform is estimated from the measured time shift. Additionally, small cracks near the hole open and close during fatigue loading, causing a modulation of the amplitude ...

Experimental verification of a Kalman filter approach for estimating the size of fastener hole fatigue cracks

Ultrasonic methods have been implemented for in situ sizing of fatigue cracks near fastener holes. These techniques, however, only provide an estimate at the time of the measurement and cannot predict the remaining life of the structure. In contrast, statistical crack propagation approaches model the expected fatigue life based on worst-case fatigue process assumptions. The authors have recently developed a Kalman filter approach for combining ultrasonic observations with crack growth laws. An ultrasonic angle-beam technique, combined with an energy-based wave propagation model, serves as the measurement model. Pariss crack growth equation acts as the system model for crack propagation. For simulated data, this approach provided more accurate crack size estimates than either the ultrasonic measurements or crack growth approach alone. Presented here are experimental results to assess the ability of the Kalman filter to provide reasonable crack size estimates.

The Effect of Transducer Placement on the Monitoring of Fatigue Cracks Emanating from Fastener Holes

In situ ultrasonic methods for structural health monitoring have the potential to provide early detection of fatigue cracks in aircraft components, specifically cracks initiating from fastener holes. A dual angle beam technique based upon an incident shear wave and a spiral creeping wave propagating around the hole has been shown to be a viable configuration for monitoring of fastener hole cracks. This technique takes advantage of the interference of the creeping wave with cracks opening and closing under load, and the ratio of received energy under a reference load to that under no load is monitored for crack detection and sizing. While prior work with this method has been very promising, transducer placement can significantly influence its performance. Here we consider the effect of transducer placement on the sensitivity of the energy ratio algorithm, using the observed location of cracking to guide placement of transducers. Ray path analysis is performed for determining the optimum transducer location...