Search

Multiple dynamic impact signals are widely used in a variety of engineering scenarios and are difficult to identify accurately and quickly due to the signal adhesion phenomenon caused by nonlinear interference. To address this problem, an intelligent algorithm combining wavelet transforms with lightweight neural networks is proposed. First, the features of multiple impact signals are analyzed by establishing a transfer model for multiple impacts in multibody dynamical systems, and interference is suppressed using wavelet transformation. Second, a lightweight neural network, i.e., fast-activated minimal gated unit (FMGU), is elaborated for multiple impact signals, which can reduce computational complexity and improve real-time performance. Third, the experimental results show that the proposed method maintains excellent feature recognition results compared to gate recurrent unit (GRU) and long short-term memory (LSTM) networks under all test datasets with varying impact speeds, while its metrics for computational complexity are 50% lower than those of the GRU and LSTM. Therefore, the proposed method is of great practical value for weak hardware application platforms that require the accurate identification of multiple dynamic impact signals in real time. [ABSTRACT FROM AUTHOR]

Sichuan-Tibet Railway spans several watersheds such as Jinsha River and Yalong River, and Potential landslides are frequent along the route, which poses serious hazards to the normal construction and operation of the railroad. The traditional time-series InSAR technology is limited by the number of images and other restrictions, and has a long solution time, making it difficult to obtain information on short-term occurrence of deformation and unable to perform wide-area potential landslides monitor quickly. In this paper, taking a geological hazard-prone area in the Jinsha River basin (or a section of the Sichuan-Tibet line) as an example, based on the Sentinel-1 satellite SAR data provided by the Copernicus program of ESA, the model corrects the interferometric superposition deformation results obtained from a small number of SAR images (less than 7), decodes the corrected rate results, and identifies a total of 13 areas where deformation obviously occurs A total of 13 typical areas with significant deformation were identified. The identified typical areas were time-series solved and their deformation was traced. The method provides a new idea for identification and monitor of Potential landslides in a wide area and further promotes the development of disaster prevention and mitigation. [ABSTRACT FROM AUTHOR]