Search

At present, in order to improve the safety performance of power battery, a safety vent is welded on the battery cover to avoid unpredictable explosions. It is vital to detect the laser welding defects on safety vent effectively for product quality. In this paper, a lightweight multiscale attention semantic segmentation algorithm with high accuracy and efficiency was proposed. We built an experimental dataset of safety vent welding defects with a total of 7263 original images, which were collected from a battery manufacturing production line. The main framework of the proposed model consists of four modules: the improved Res2Net serving as the feature extraction sub-module, an attention mechanism, a localization block and a boundary anti-aliasing module. This architecture can segment defects of different sizes and shapes in real-time and get more refined segmentation results simultaneously. To evaluate the method, experiments concerning mean IOU and pixel accuracy were conducted, and an average validation accuracy of 99.4% and the mean IOU of 84.67% were achieved respectively. Furthermore, comparison experiments using some outstanding algorithms on safety vent’s welding defects test dataset were performed. It proves that our method achieved the best performance in terms of model size, computational complexity, efficiency and detection accuracy. Specifically, the model size is only 3.8 MB, and the frames per second (FPS) is 132.3. In brief, the proposed model is suitable for laser welding quality detection on safety vent in an industrial environment. Additionally, our study can provide a reference for designing relevant defect detection tasks using semantic segmentation method.

In optical interconnection field, inconsistent coupling behaviors of TE and TM modes may cause abnormal operation. Therefore, it is very significant to implement a polarization-independent directional coupler. In this paper, a hybrid-plasmonic-waveguide-based polarization-independent directional coupler has been proposed. Since the energy of the hybrid plasmonic waveguides for TE and TM modes is distributed in different layers, we manage to achieve polarization independence by adjusting the material properties and dimensions of the corresponding layer. We optimize the parameters of our proposed directional coupler such as radius of silicon waveguide layer, height of silica waveguide layer, and the distance between the two waveguides, etc. Then performance of the directional coupler has been evaluated. It is worth mentioning that the length of the coupling section is only 4.25μm. Meanwhile, the polarization-dependent loss is only 0.393 dB, and the maximum coupling efficiency of TE and TM modes can reach 86.4% and 94.6%, respectively. Besides, the coupling efficiency of TM mode remains above 90% over the entire C-band, while the coupling efficiency also keeps at least 80% for TE mode. Finally, the manufacture process for the proposed directional coupler has been discussed. In brief, the improved polarization-independent directional coupler features small size, low energy loss, good polarization-independent characteristics, and wavelength insensitivity simultaneously. Compared to the other counterparts ever proposed, our proposed coupler can provide a perfect trade-off between device size and loss, which shows important potential applications in the fields of PIC and optical interconnection.

Unmanned aerial vehicle (UAV) nowadays is a promising technology for boosting wireless connectivity, especially for upcoming 5G wireless networks. Due to its aerial feature, potential mobility and flexible deployment, the UAV can provide service in some specific scenarios such as post-disaster network recovery or no-infrastructure terrains. However, the problem of interference management between deployed UAV and underlaying heterogeneous networks, which guarantees the quality of service, is still a challenging task. In this paper, we address the outage probability for UAV connected users and device-to-device (D2D) receivers simultaneously operating in D2D underlaying NOMA UAV-assisted networks, and we derive its closed-form expressions, which turns out to be very accurate. We also solve the power control optimization problem via relaxing the non-convex problem into solving successive low-complexity linear programs to obtain a sub-optimal solution to the problem. The simulation results in confirm the superiority of the proposed approach in the terms of computational complexity and its compromise in the terms of sum rate.