Search

Machine learning for nondestructive evaluation (NDE) has the potential to bring significant improvements in defect characterization accuracy due to its effectiveness in pattern recognition problems. However, the application of modern machine learning methods to NDE has been obstructed by the scarcity of real defect data to train on. This article demonstrates how an efficient, hybrid finite element (FE) and ray-based simulation can be used to train a convolutional neural network (CNN) to characterize real defects. To demonstrate this methodology, an inline pipe inspection application is considered. This uses four plane wave images from two arrays and is applied to the characterization of cracks of length 1-5 mm and inclined at angles of up to 20° from the vertical. A standard image-based sizing technique, the 6-dB drop method, is used as a comparison point. For the 6-dB drop method, the average absolute error in length and angle prediction is ±1.1 mm and ±8.6°, respectively, while the CNN is almost four times more accurate at ±0.29 mm and ±2.9°. To demonstrate the adaptability of the deep learning approach, an error in sound speed estimation is included in the training and test set. With a maximum error of 10% in shear and longitudinal sound speed, the 6-dB drop method has an average error of ±1.5 mmm and ±12°, while the CNN has ±0.45 mm and ±3.0°. This demonstrates far superior crack characterization accuracy by using deep learning rather than traditional image-based sizing.

Plane wave imaging (PWI) is an ultrasonic array imaging technique used in nondestructive testing, which has been shown to yield high resolution with few transmissions. Only a few published examples are available of PWI of components with nonplanar surfaces in immersion. In these cases, inspections were performed by adapting the transmission delays in order to produce a plane wave inside the component. This adaptation requires prior knowledge of the component geometry and position relative to the array. This article proposes a new implementation, termed PWI adapted in postprocessing (PWAPP), which has no such requirement. In PWAPP, the array emits a plane wave as in conventional PWI. The captured data are input into two postprocessing stages. The first reconstructs the surface of the component; the latter images inside of it by adapting the delays to the distortion of the plane waves upon refraction at the reconstructed surface. Simulation and experimental data are produced from an immersed sample with a concave front surface and artificial defects. These are processed with conventional and surface corrected PWI. Both algorithms involving surface adaptation produced nearly equivalent results from the simulated data, and both outperform the nonadapted one. Experimentally, all defects are imaged with a signal-to-noise ratio (SNR) of at least 31.8 and 33.5 dB for, respectively, PWAPP and PWI adapted in transmission but only 20.5 dB for conventional PWI. In the cases considered, reducing the number of transmissions below the number of array elements shows that PWAPP maintains its high SNR performance down to the number of firings equivalent to a quarter of the array elements. Finally, experimental data from a more complex surface specimen are processed with PWAPP resulting in detection of all scatterers and producing SNR comparable to that of the total focusing method.

The thesis describes experiments on isotope effects on steric and other interactions that may be used to test, for example, the validity of empirical potential functions for non-bonded interactions involving hydrogen. The earlier serendipitous observation of shifts in π → π* UV absorption bands on deuteriation is exploited as a possible probe for examination of the electronic and geometric changes accompanying deuteriation. The effect of deuterium substitution on the free energy differences of conformational equilibria with hydrogen in two possible environments is studied for a series of 2-substituted-1,3-dioxanes and 3-phenyl-3,5,5-trimethylcyclohexanone. The time-averaged NMR chemical shifts of suitable protons in the deuteriated and undeuteriated molecules are used to indicate the positions of the equilibria in several solvents, and model compounds are used to provide shift values from which free energy differences for the equilibrating systems may be calculated with the same systematic errors in both the undeuteriated and deuteriated systems. Possible sources of error are discussed, and the largest is shown to be the standard deviation of measurement of the time-averaged NMR chemical shift differences, leading to errors of 1 to 2 cal/mole in isotope effects from -2 to +16 cal/mole on strain energies ~ 4 kcal/mole. The measured effects were not always of a sign compatible with a reduction of "size" on changing from an undeuteriated to a deuteriated group. Preliminary investigations of systems for which isotope effects on intermolecular proton exchange may be measured, are reported, and isotope effects on the free energy differences between the two hydroxyl groups of partially deuteriated 3,5-di-t-buryl-2,6-dihydroxyacetophenone and of partially deuteriated 3,5-di-t-butyl-2,6-dihydroxybenzaldehyde O-methyloxime of 50 and 20 cal/mole respectively, protium being favoured in the intramolecularly hydrogen bonded group, are measured by the NMR chemical shifts of hydroxyl proton signals. These chemical shift values are used in conjunction with the hydroxyl proton chemical shifts of suitable models to determine the quoted isotopic changes in free energy differences. Isotopic changes in the intensities of UV absorption bands are measured for cis- and trans-3,5-di-t-butylcyclohexane-1,3-diones and interpreted in terms of changes in the free energy difference between keto-enol and diketo tautomers. No such change was detected on deuteriation of acetylacetone, contrary to the results reported by Allred and Thompson. Isotopic shifts in the frequency position of the centroids of UV bands are measured by "difference" spectrometry, a technique whose applicability and accuracy is illustrated by the reproducibility and insensitivity to calibration errors of shifts in the centroids of the IR carbonyl stretching bands between a series of cyclohexanone derivatives measured in the same way. The UV band shifts are interpreted as isotopic changes of differences in the potential energy between the ground state and geometrically different excited states for the enol forms of the above diketones and a series of 4-aminopyridine derivatives. All the observed effects are interpreted in terms of the reduction of zero-point vibrational energy accompanying substitution of deuterium for hydrogen. However the effect of the smaller "size" of deuteriated groups on steric strain energies appears to be accompanied by isotope effects on other contributions, for example those caused by bond polarities, to non-bonded interactions which may enhance or diminish the "size" effect.