Transmission line detection in aerial images: An instance segmentation approach based on multitask neural networks

Camera-based transmission line detection (TLD) is a fundamental and crucial task for automatically patrolling powerlines by aircraft. Motivated by instance segmentation, a TLD algorithm is proposed in this paper with a novel deep neural network, i.e., CableNet. The network structure is designed based on fully convolutional networks (FCNs) with two major improvements, considering the specific appearance characteristics of transmission lines. First, overlaying dilated convolutional layers and spatial convolutional layers are configured to better represent continuous long and thin cable shapes. Second, two branches of outputs are arranged to generate multidimensional feature maps for instance segmentation. Thus, cable pixels can be detected and assigned cable IDs simultaneously. Multiple experiments are conducted on aerial images, and the results show that the proposed algorithm obtains reliable detection performance and is superior to traditional TLD methods. Meanwhile, segmented pixels can be accurately identified as cable instances, contributing to line fitting for further applications.     

Surface modification of cellulose nanocrystals towards new materials development

Cellulose nanocrystals (CNCs) are a kind of sustainable nanoparticle from biomass, which are widely used as reinforcing filler and assembly building block for high-performance composites and function materials including biomaterial, optics, and so forth. Here, their unique advantages in material applications were reviewed based on their rod-like morphology, crystalline structure, dimension-related effects, and multi-level order structure. Then, we focused on the molecular engineering of CNCs, including the structure and physicochemical properties of their surface, along with surface modification methods and steric effects. We further discussed the performance-improvement and functionalization methods based on multi-component complex systems, together with the effects of surface molecular engineering on the performance and functions. Meanwhile, methods of optimizing orientation in uniaxial arrays were discussed along with those of enhancing photoluminescence efficiency via surface chemical modification and substance coordination. In the end, we prospected the design, development, and construction methods of new CNCs materials.     

超宽带大功率合成器设计北大核心CSCD

在80 MHz~1 GHz频段,单个功率管输出功率能达到100 W以上,为研制输出功率400 W的功率放大器,文中设计了四路功率合成器。该合成器需要实现功率容量大、工作频带宽、体积小的设计目标。在功率容量方面,文中采用悬置带状线结构,其功率容量远远大于微带线结构;在工作频带方面,采用切比雪夫九节阻抗变换器,将工作带宽拓宽为80 MHz~1 GHz;在体积方面,文中合成器的功率合成部分采用Y型节级联实现四路功率合成,阻抗变换部分采用切比雪夫阻抗变换器进行阻抗变换,该结构相较于磁环巴伦功率合成器,不但具有损耗小、平坦度高的优点,而且通过将阻抗变换器设计成曲折的形状,进一步缩小了合成器体积。仿真与实测结果显示该合成器在80 MHz~1 GHz范围内还具有较高的平坦度,合成效率可达90%以上。     

One-step to prepare high-performance gas diffusion layer (GDL) with three different functional layers for proton exchange membrane fuel cells (PEMFCs)

Gas diffusion layer (GDL) is one of the most important components of fuel cells. In order to improve the fuel cell performance, GDL has developed from single layer to dual layers, and then to multiple layers. However, dual or multi layers in GDL are usually prepared by layer-by-layer methods, which cost too much time, energy, and resources. In this work, we successfully developed a facile one-step method to prepare a GDL with three functional layers by utilizing the different sedimentation rates and filtration rates of short carbon fiber (CF) and carbon nanotube (CNT). The treatment temperature for this GDL is much lower than that of traditional method. The thickness of the GDL can be effectively controlled from as thin as 50 μm to more than 200 μm by simply adjusting the content of CF. The GDL with high flexibility is suitable to develop high performance flexible electronics. The fuel cell with the GDL has the maximum power density 1021 mW cm^?2, which shows 19% improvement comparing to the conventional one. Therefore, this work breaks the traditional concept that GDL for fuel cells only can be prepared by very complex and high-cost procedure.     

Recent progress of electrocatalysts for hydrogen proton exchange membrane fuel cells

Due to stringent environmental regulations and the limited resources of fossil-based fuels, there is an urgent demand for clean and eco-friendly energy conversion devices. These criteria appear to be met by hydrogen proton exchange membrane fuel cells (PEMFCs). PEMFCs have attracted tremendous attention on account of their excellent performance with tunable operability and good portability. Nonetheless, their practical applications are hugely influenced by the scarcity and high cost of platinum (Pt) used as electrocatalysts at both cathode and anode. Pt is also susceptible to easy catalyst poisoning. Herein, this paper reviews the progress of the research regarding the development of electrocatalysts practically used in hydrogen PEMFCs, where the corner-stone reactions are cathodic oxygen reduction reaction (ORR) and anodic hydrogen oxidation reaction (HOR). To reduce the costs of PEMFCs, lessening or eliminating the use of Pt is of prime importance. For current and forthcoming laboratory/large-scale PEMFCs, there is much interest in developing substitute catalysts based on cheaper materials. As such are non-platinum (non-Pt), non-platinum group metals (non-PGMs), metal oxides, and non-metal electrocatalysts. Hence, high-performance, state-of-the-art, and novel structured electrocatalysts as replacements for Pt are needed.     

基于人工神经网络的智能化架空线路应变快速解调方法北大核心

为了提高智能化光纤复合架空线路态势感知的实时性,将人工神经网络方法应用于光纤沿线应变解调,确定了神经网络的结构。编程实现了基于洛伦兹模型的最小二乘谱拟合方法和神经网络方法,采用不同信噪比和布里渊频移的布里渊谱训练神经网络,将它们应用于某光纤复合架空线路沿线光纤应变的测量,从不同角度比较了两种方法的计算结果。计算结果表明,神经网络方法能有效获得光纤沿线的布里渊频移进而获得应变,具有与谱拟合方法相似的准确性,但应变解调时间仅约为谱拟合方法的1/20000。研究结果为提高智能光纤复合架空线路态势感知的实时性提供了参考。     

离线式带钢双面翻转检查站方案设计及实践

针对某酸轧机组检查站穿带困难、检查效率低、产生废品多、故障率高的问题,开发了一种离线式带钢双面检查站。该检查站新增了带钢翻转装置,将在入口分切剪处完成定尺剪切的带钢试样由两侧压板夹紧,中间可升降皮带机落下,翻转装置将带钢试样张紧,并翻转90°或者180°,实现带钢的双面检查。皮带机两侧布置有可升降、可横移的活动操作平台,检测人员站在平台上对带钢进行打磨检查,可同时对带钢上下表面进行检测。由于缩短了带钢的传输距离,有效减少了其他擦划伤的影响,提高了带钢表面质量检测准确性。     

Triggering the aqueous interparticle association of γ‒Al2O3 hierarchical assemblies using divalent cations and cellulose nanofibers

Understanding aqueous dispersion, rheological properties and colloidal stabilisation mechanisms of hierarchically assembled ceramic powders is important for progress in the fields of catalysis, separation and/or adsorption. The present study was designed to evaluate the rheological and sedimentation behaviour of highly loaded aqueous suspensions (up to φ_A = 0.126) containing AlN-powder-hydrolysis-derived, micron-sized, mesoporous, gamma alumina (MA) particulates with a high surface area (～180 m²/g) dispersed with sodium polyacrylate (NaPAA). The as-prepared suspensions were prone to sedimentation and segregation. However, when divalent cations (Mg²⁺, Ca²⁺) or cellulose nanofibers were added, the formation of interparticle association networks in the aqueous suspensions containing MA particles was triggered, facilitating their long-term resistance to sedimentation lasting more than 12 weeks.     

Optical design of the Lyman alpha based beam emission spectroscopy(LAB)diagnostic on the HL-2A tokamak

In this article, we present the optical design of a novel diagnostic on the HL-2 A tokamak, i.e. the20-channel edge Lyman-alpha beam emission spectroscopy, which is a promising solution for edge density turbulence research on tokamaks, as it offers the possibility of density fluctuation measurement with a 3.3 mm spatial resolution while maintains a high temporal resolution of1 μs. The optical path, including the reflective collection optics, the high-dispersion spectrometer, and the linear detector array, is carefully optimized to obtain a good image quality and a high throughput. The maximum root mean square radius of the collection optics is 64 μm.The detected photon flux is estimated to be about 10~(11) photons/s/channel.     

基于阶段注意力机制的电力线提取算法

电力系统维护是电力系统稳定运行的重要保障，应用智能算法的无人机电力巡检则为电力系统维护提供便捷。电力线提取是自主电力巡检以及保障飞行器低空飞行安全的关键技术，结合深度学习理论进行电力线提取是电力巡检的重要突破点。本文将深度学习方法用于电力线提取任务，结合电力线图像特点嵌入改进的图像输入策略和注意力模块，提出一种基于阶段注意力机制的电力线提取模型（SA-Unet）。本文提出的SA-Unet模型编码阶段采用阶段输入融合策略（Stage input fusion strategy， SIFS），充分利用图像的多尺度信息减少空间位置信息丢失。解码阶段通过嵌入阶段注意力模块（Stage attention module，SAM）聚焦电力线特征，从大量信息中快速筛选出高价值信息。实验结果表明，该方法在复杂背景的多场景中具有良好的性能。