Noncontacting, laser-based resonant ultrasound spectroscopy (RUS) was applied to characterize the microstructure of a polycrystalline sample of high purity copper. The frequencies and shapes of 40 of the first 50 resonant vibrational modes were determined. The sample's elastic constants, used for theoretical prediction, were estimated using electron backscatter diffraction data to form a polycrystalline average. The difference in mode frequency between theory and experiment averages 0.7% per mode. The close agreement demonstrates that, using standard metallurgical imaging as a guide, laser-based RUS is a promising approach to characterizing material microstructure. In addition to peak location, the Q of the resonant peaks was also examined. The average Q of the lasergenerated and laser-detected resonant ultrasound spectrum was 30% higher than a spectrum produced employing a piezoelectric transducer pair for excitation and detection.

No abstract

Articles you may be interested inSelective laser melting of aluminum die-cast alloy-Correlations between process parameters, solidification conditions, and resulting mechanical properties J. Laser Appl. 27, S29205 (2015) In situ laser-based resonant ultrasound spectroscopy is used to characterize the development of a recrystallized microstructure in a high purity copper sample. The modal shapes, used for mode identification, of several resonant modes are determined before and after annealing by raster scanning the laser interferometric probe. This information is used to isolate the motion of individual modes during high temperature annealing. The evolution of a particular mode during annealing is examined in detail. During recrystallization, the center frequency of this mode shifts by approximately 20% of the original value. Using electron backscatter data it is shown that the majority of this shift is due to changes in the polycrystal average elastic stiffness tensor, driven by changes in texture, and that changes in dislocation density and pinning length are secondary influences.

Industrial Control Systems (ICS) are used to control physical processes in critical infrastructure. These systems are used in a wide variety of operations such as water treatment, power generation and distribution, and manufacturing. While the safety and security of these systems are of serious concern, recent reports have shown an increase in targeted attacks aimed at manipulating physical processes to cause catastrophic consequences. This trend emphasizes the need for algorithms and tools that provide resilient and smart attack detection mechanisms to protect ICS. In this paper, we propose an anomaly detection framework for ICS based on a deep neural network. The proposed methodology uses dilated convolution and long short-term memory (LSTM) layers to learn temporal as well as long term dependencies within sensor and actuator data in an ICS. The sensor/actuator data are passed through a unique feature engineering pipeline where wavelet transformation is applied to the sensor signals to extract features that are fed into the model. Additionally, this paper explores four variations of supervised deep learning models, as well as an unsupervised support vector machine (SVM) model for this problem. The proposed framework is validated on Secure Water Treatment testbed results. This framework detects more attacks in a shorter period of time than previously published methods.

Two copper specimens with distinct grain microstructures are investigated using laser resonant ultrasound spectroscopy (LRUS). One consists of randomly oriented crystallites and exhibits isotropic elastic behavior (two elastic constants), and the other has been highly textured by rolling and exhibits anisotropic elastic behavior (three elastic constants). The elastic constants are measured using electron backscatter diffraction, LRUS, and time domain laser ultrasound (LU). The elastic constants of the isotropic sample obtained via electron backscatter diffraction (EBSD), LU, and LRUS agree closely. However, for the anisotropic sample, there is considerable disagreement between results obtained using LRUS and results obtained using LU and EBSD. Analysis reveals that increasing the dimensionality of the modulus space leads to a questions of whether the LRUS results are unique to within experimental error. The consequence is that for anisotropic materials, small measurement uncertainties can lead to large uncertainties in the measured elastic constants. This observation has important implications for the use of LRUS to measure the elastic constants of thin texture samples.