车牌识别嵌入式系统的软硬件设计

作为交通信息服务系统的重要手段,车牌识别技术避免了人工操作存在的各种弊端。文中设计了车牌识别嵌入式系统的软硬件,该系统的整体框架主要包含3个模块,分别是OV7670摄像头、STM32单片机和TFT显示屏。为满足任务需求,设计了最小系统和电源模块等硬件;为解决传统投影法误差较大的问题,该系统在字符分割前会先对车牌区域进行识别,并采用上下投影的方式在有效区域内进行字符分割。实验表明,该系统可以实现车牌识别功能,且精确度较高。     

In Situ Polymerizing Internal Encapsulation Strategy Enables Stable Perovskite Solar Cells toward Lead Leakage Suppression

Despite the outstanding power conversion efficiency (PCE) of perovskite solar cells (PSCs) achieved over the years, unsatisfactory stability and lead toxicity remain obstacles that limit their competitiveness and large-scale practical deployment. In this study, in situ polymerizing internal encapsulation (IPIE) is developed as a holistic approach to overcome these challenges. The uniform polymer internal package layer constructed by thermally triggered cross-linkable monomers not only solidifies the ionic perovskite crystalline by strong electron-withdrawing/donating chemical sites, but also acts as a water penetration and ion migration barrier to prolong shelf life under harsh environments. The optimized MAPbI₃ and FAPbI₃ devices with IPIE treatment yield impressive efficiencies of 22.29% and 24.12%, respectively, accompanied by remarkably enhanced environmental and mechanical stabilities. In addition, toxic water-soluble lead leakage is minimized by the synergetic effect of the physical encapsulation wall and chemical chelation conferred by the IPIE. Hence, this strategy provides a feasible route for preparing efficient, stable, and eco-friendly PSCs.     

Adaptive Clustered Federated Learning for Heterogeneous Data in Edge Computing

Mobile Networks and Applications - Although federated learning has been widely used in collaborative training of machine learning models, its practical uses are still challenged by heterogeneous...     

Low-rank tensor completion via combined Tucker and Tensor Train for color image recovery

Applied Intelligence - In recent years, low-rank tensor completion has been widely used in color image recovery. Tensor Train (TT), as a balanced tensor rank minimization method, has achieved good...     

发动机进气系统高通量富氧膜的研制

采用交联改性的方法对较低分子量的乙烯基硅橡胶进行了改性研究,制备了高通量的富氧膜,结合单因素实验法和Design-expert正交实验设计,考察了原料配比、交联反应时间、固化温度和固化时间等因素对富氧膜性能的影响,确定了最佳富氧性能膜的制备条件:原料配比4.3,反应时间2.15h,固化温度86.25℃,固化时间1.75h;获得了富氧浓度为28.68%,透气量为4696.33 barrer的富氧膜,该膜的富氧浓度与常规商品化膜相当,透气量为常规商品化膜的2~4倍,有应用于发动机富氧进气系统的前景。通过扫描电镜检测、机械性能检测等手段对复合膜进行了进一步表征。     

MHSACAE-CNN在噪声下的电机轴承故障诊断

电机的运行情况复杂,实际运行工况下会有大量的噪声,导致其轴承故障诊断精度下降。为了改善这一问题,提出了一种基于多头自注意力机制的一维全卷积自编码网络（One-dimensional Fully Convolutional Autoencoding Network Basedon Multi-head Self-attention,MHSACAE）与卷积神经网络（Convolutional Neural Network,CNN）结合的轴承故障诊断方法。该方法先采用MHSACAE网络进行降噪,再通过CNN进行故障诊断。其中MHSACAE去噪网络采用无监督训练的方式,充分考虑了实际工况和序列数据内在联系,在实现对噪声的滤除效果的同时,最大限度地保留下了原始的故障信息,使得CNN可以实现在噪声情况下对电机轴承故障的高精度诊断。通过与其他轴承故障诊断方法在噪声情况下进行对比,证明提出的方法具有更好的效果。     

Fabrication of Titanium Aluminides via Powder-Cored Wire Arc Additive Manufacturing: Microstructural and Mechanical Characterization

Herein, an innovative powder-cored wire arc additive manufacturing (PC-WAAM) process is proposed to fabricate γ-TiAl thin-walled intermetallic alloy. The metallography, phase composition, and mechanical properties at different thin-wall locations are characterized. The results show that the alternatively distributed layer-like microstructure composed of α₂ (Ti₃Al) and γ (TiAl) phases is obtained along the building direction. The content of α₂ phase exhibits the tendency of decreasing from the bottom to top region. This unique microstructure characteristic is closely related to the typical thermal cycling history during deposition. Moreover, the tensile strength and microhardness of the top region are lower than the middle and bottom region. In general, the current PC-WAAM technique shows promising capability of fabricating γ-TiAl intermetallic alloy with low cost. This work becomes a valuable reference for understanding the evolution mechanism of microstructure and paves the way for the flexible and customized additive manufacturing of γ-TiAl alloy.