Jian Feng

The fundamental shear horizontal (SH0) wave is widely used for structural health monitoring due to its high detection sensitivity and long propagation range. However, the typical SH0 waves face a trade-off between spatial resolution and signal-to-noise ratio (SNR) when encountering two or more adjacent micro defects. Existing broadband excitation techniques and passive sidelobe suppression methods alone cannot fully overcome the abovementioned challenges. To address these issues, this article proposes an orthogonal complementary spatiotemporal modulation method. First, a spatiotemporal modulation strategy is formulated to generate SH0 waves characterized by a narrow, high-amplitude main lobe, affording a substantial enhancement in spatial resolution with no concomitant degradation in SNR. Second, an active sidelobe suppression method is developed to identify micro defect echoes from macro defect signals by exploiting the complementary radiation characteristics of multiarray acoustic fields. Third, an optimized excitation-reception strategy is presented to realize unilateral reception, thereby further elevating the SNR of micro defect responses. Experimental validation demonstrates that the proposed technique effectively distinguishes a 0.3 mm × 1 mm micro defect located one wavelength away from a macro defect, offering a promising solution for the accurate localization and characterization of submillimeter defect clusters.

Non-intrusive load monitoring (NILM) remains challenging due to the non-stationary and multi-scale characteristics of aggregated power signals. To address these issues, this paper proposes a time-frequency guided feature learning framework that integrates wavelet-based signal decomposition with multi-scale temporal modeling. A wavelet packet transform (WPT) is first applied to obtain a time-frequency representation of the input signal. An adaptive wavelet weighting (AWW) mechanism is then introduced to selectively emphasize informative frequency subbands. To further capture temporal patterns at different resolutions, a cross-scale convolutional pyramid (CSCP) is designed to fuse multi-scale local features through parallel dilated convolutions and attention-based weighting. The resulting features are finally aggregated using a global temporal modeling module for appliance-level power disaggregation. Experimental results on three public datasets show that the proposed method achieves competitive accuracy compared with state-of-the-art methods.