Ultrasonics最新文献

Guided wave ultrasonic testing (GWUT) in industrial environments is often limited by low signal-to-noise ratio (SNR), which reduces defect detectability. This study proposes a knowledge-guided framework that combines synthetic data generation with a tailored denoising network. From a single reference acquisition, paired clean and noisy signals are constructed using dual-Gaussian echo modeling and composite noise synthesis based on measured spectra. A Wavelet-Initialized Attention U-Net is developed with wavelet-informed kernels, a dual-decoder structure, and an attention bottleneck for efficient temporal integration. Experiments on two representative GWUT systems, a railway switch rail monitoring setup and a storage tank wall inspection robot, show that the proposed framework achieves up to 29.7 dB ROI-based SNR improvement on synthetic data, and substantial CNR improvement on real signals accompanied by a marked reduction of false detections (FP/FN), outperforming classical and deep learning baselines. The method also achieves real-time inference and efficient data generation with moderate computational cost. These results indicate that physics-guided synthesis combined with a tailored network provides a practical solution for GWUT denoising and supports reliable defect detection in industrial applications.

Conventional sandwich-type ultrasonic motors suffer from structural complexity, limited design flexibility, and inefficient assembly when configured for dual-rotor operation, which restricts their application scope. To address these challenges, this study proposes a novel piezoelectric actuation mechanism, an expandable design scheme, and a cooperative assembly method tailored for sandwich-type dual-rotor ultrasonic motors. The principle of longitudinal-bending modal coupling is employed to replace the commonly adopted bending-bending mechanism in dual-rotor actuation, thereby simplifying the coupling elements. An expandable design method is established, enabling rapid iteration of transducer count, ring structure dimensions, and key vibration modes according to specific requirements, enhancing prototype development flexibility for engineering applications. A case study is presented accordingly, its core structural advantage lies in a synergistic assembly system formed by three components: the frame, prestressed nuts, and ring-horn composite structure, enabling efficient and stable multi-transducer cooperative assembly through single-step torque loading. This approach avoids the inefficiency and low reliability associated with conventional individual transducer assembly via standard bolts. The validity of the driving principle is verified through finite element analysis, and the structural parameters are optimized. The feasibility of the assembly strategy and design method is validated through impedance-phase and vibration measurements on the prototype. Mechanical performance tests show the prototype's output torque reaches 1.5 N·m (600 V in peak-to-peak value) with reliable continuous operation. Notably, the prototype exhibits exceptional stepping capability, achieving single-step angular displacements of 0.065/0.114μrad in single-/dual-rotor configurations. The study expands the design paradigm for sandwich-type dual-rotor ultrasonic motors, providing a systematic solution for engineering applications.

Addressing the long-tail problem (LTP) is critical when applying deep learning (DL) to ultrasonic testing, as defective samples often lead to poor testing performance. This study addresses the LTP in stress-strain curve prediction using ultrasound by applying a Value Imputation and Mask Estimation (VIME)-based self-supervised learning (SSL) framework. Using 816 aluminum alloy samples, including low yield strength (YS) cases (100-200 MPa) that trigger LTP, the baseline model performed well overall but degraded sharply on LTP data (mean absolute percentage error (MAPE): 10% for non-LTP vs. 26% for LTP). VIME-SSL reduced the MAPE to 9.4% and 21%, respectively, with greater relative improvement for LTP cases. Notably, frequency-domain signals containing fundamental and second harmonic components were found to be especially effective for VIME-SSL in addressing the LTP. This finding was substantiated by separate ultrasonic measurements of attenuation and nonlinearity. Overall, this study demonstrates VIME-SSL as a promising approach for improving DL-based ultrasonic testing on rare or anomalous samples.

Osteoporosis is a public health concern strongly associated with menopause and aging. In this study, ovariectomized (OVX) and spontaneously menopausal (SM) C57BL/6J mice were used to experimentally investigate the effects of menopause and aging on the skull bone (unloaded bone). Ultrasonic longitudinal wave velocities and volumetric bone mineral density (v.BMD) were measured using micro-Brillouin scattering (μ-BS) and X-ray micro-computed tomography (μ-CT). The experimental findings revealed that v.BMD was significantly lower in OVX and SM mice than in Sham-operated (control) mice. The measured wave velocity in the anteroposterior direction of tibiae showed a significant reduction in SM mice, whereas no significant difference was found between Sham and OVX mice. A positive correlation was found between the square of wave velocity and v.BMD (r = 0.65). The skull exhibits complex elastic characteristics owing to the lack of mechanical loading. In the load-bearing bones (tibia), the direction of loading corresponded to the direction of the maximum wave velocity. Conversely, the directions of the maximum and minimum wave velocities in the skull varied depending on the sample. However, SM skull samples exhibited lower wave velocities.