摘要：Investigating highly effective electrocatalysts for high-temperature proton exchange membrane fuel cells(HT-PEMFC)requires the resistance to phosphate acid(PA)poisoning at cathodic oxygen reduction reaction(ORR).Recent advancements in catalysts have focused on alleviating phosphoric anion adsorption on Pt-based catalysts with modified electronic structure or catalytic interface and developing Fe-N-C based catalysts with immunity of PA ***-N-C-based catalysts have emerged as promising alternatives to Pt-based catalysts,offering significant potential to overcome the characteristic adsorption of phosphate anion on *** overview of these developments provides insights into catalytic mechanisms and facilitates the design of more efficient *** review begins with an exploration of basic poisoning principles,followed by a critical summary of characterization techniques employed to identified the underlying mechanism of poisoning *** is then directed to endeavors aimed at enhancing the HT-PEMFC performance by well-designed ***,the opportunities and challenges in developing the anti-PA poisoning strategy and practical HT-PEMFC is *** these discussions,a comprehensive understanding of PA-poisoning bottlenecks and inspire future research directions is aim to provided.

摘要：Structural design optimization has always been a topic of concern in industry because good design can improve the safety and economic efficiency of structures during their service *** the appropriate optimization algorithm is the key to solving structural optimal design *** this study,a new global optimization idea is proposed and named the moving baseline strategy.A baseline is initially set and will be repeatedly moving upward or downward to approach the optimal *** proposed strategy is a simple but effective,general,and stable algorithm that can be used to solve constrained and unconstrained structural optimization *** from traditional gradient-based,stochastic and heuristic algorithms,the developed algorithm provides a completely new idea to solve global or local optimization *** unconstrained and constrained numerical benchmark examples are used to test the proposed *** addition,structural optimal design problems of a ten-bar planar truss structure and a hypersonic wing structure(X-37B)are utilized to verify the effectiveness of the developed strategy in addressing structural design optimization problems in engineering.

摘要：In order to meet the demands for high transmission rates and high service quality in broadband wireless communication systems, orthogonal frequency division multiplexing （OFDM） has been adopted in some standards. However, the inter-block interference （IBI） and inter-carrier interference （ICI） in an OFDM system affect the performance. To mitigate IBI and ICI, some pre-processing approaches have been proposed based on full channel state information （CSI）, which improved the system performance. A pre-processing filter based on partial CSI at the transmitter is designed and investigated. The filter coefficient is given by the optimization processing, the symbol error rate （SER） is tested, and the computation complexity of the proposed scheme is analyzed. Computer simulation results show that the proposed pre-processing filter can effectively mitigate IBI and ICI and the performance can be improved. Compared with pre-processing approaches at the transmitter based on full CSI, the proposed scheme has high spectral efficiency, limited CSI feedback and low computation complexity.

摘要：With the growing complexity of engineering structures and the severity of their service environments,the uncertainties related to material properties,external loads,boundary conditions,measurement noises,etc.,have *** greatly influences structural safety,and hence,it is necessary to completely understand all the uncertainties involved when tackling analysis and design issues[1].

摘要：Floorplan is an important process whose quality determines the timing closure in integrated circuit(IC)physical *** generating a floorplan with satisfying timing result is time-consuming because much time is spent on the generation-evaluation *** machine learning to the floorplan stage is a potential method to accelerate the floorplan ***,there exist two challenges which are selecting proper features and achieving a satisfying model *** this paper,we propose a machine learning framework for floorplan acceleration with feature selection and model stacking to cope with the challenges,targeting to reduce time and effort in integrated circuit physical ***,the proposed framework supports predicting post-route slack of static random-access memory(SRAM)in the early floorplan ***,we introduce a feature selection method to rank and select important *** both feature importance and model accuracy,we reduce the number of features from 27 to 15(44%reduction),which can simplify the dataset and help educate novice ***,we build a stacking model by combining different kinds of models to improve *** 28 nm technology,we achieve the mean absolute error of slacks less than 23.03 ps and effectively accelerate the floorplan process by reducing evaluation time from 8 hours to less than 60 *** on our proposed framework,we can do design space exploration for thousands of locations of SRAM instances in few seconds,much more quickly than the traditional *** practical application,we improve the slacks of SRAMs more than 75.5 ps(177%improvement)on average than the initial design.

摘要：Serial robots are used to handle workpieces with large dimensions, and calibrating kinematic parameters is one of the most efficient ways to upgrade their accuracy. Many models are set up to investigate how many kinematic parameters can be identified to meet the minimal principle, but the base frame and the kinematic parameter are indistinctly calibrated in a one-step way. A two-step method of calibrating kinematic parameters is proposed to improve the accuracy of the robot＇s base frame and kinematic parameters. The forward kinematics described with respect to the measuring coordinate frame are established based on the product- of-exponential （POE） formula. In the first step the robot＇s base coordinate frame is calibrated by the unit quaternion form. The errors of both the robot＇s reference configuration and the base coordinate frame＇s pose are equivalently transformed to the zero-position errors of the robot＇s joints. The simplified model of the robot＇s positioning error is established in second-power explicit expressions. Then the identification model is finished by the least square method, requiring measuring position coordinates only. The complete subtasks of calibrating the robot＇ s 39 kinematic parameters are finished in the second step. It＇s proved by a group of calibration experiments that by the proposed two-step calibration method the average absolute accuracy of industrial robots is updated to 0.23 mm. This paper presents that the robot＇s base frame should be calibrated before its kinematic parameters in order to upgrade its absolute positioning accuracy.

摘要：Low Earth Orbit(LEO)satellite for navigation augmentation applications can significantly reduce the precise positioning convergence time and attract increasing attention recently.A few LEO Navigation Augmentation(LEO-NA)constellations have been proposed,while corresponding constellation design methodologies have not been systematically *** LEO-NA constellation generally consists of a huge number of LEO satellites and it strives for multiple optimization *** is essentially different from the communication constellation or earth observing constellation design *** this study,we modeled the LEO-NA constellation design problem as a multi-objective optimization problem and solve this problem with the MultiObjective Particle Swarm Optimization(MOPSO)*** objectives are used to strive for the best tradeoff between the augmentation performance and deployment efficiency,namely the Position Dilution of Precision(PDOP),visible LEO satellites and the orbit altitude.A fuzzy set approach is used to select the final constellation from a set of Pareto optimal solutions given by the MOPSO *** evaluate the performance of the optimized constellation,we tested two constellations with 144 and 288 satellites and each constellation has three optimization schemes:the polar constellation,the single-layer constellation and the two-layer *** results indicate that the optimized two-layer constellation achieves the best global coverage and is followed by the single-layer *** MOPSO algorithm can help to improve the constellation design and is suitable for solving the LEO-NA constellation design problem.

摘要：工业互联网平台被公认为智能制造的必要推动者,使物理制造资源得以虚拟化,并允许资源以服务的形式进行协作。作为平台的核心功能,制造服务协作优化致力于为制造任务提供高质量的服务协作解决方案。这种优化与任务的功能和数量要求密不可分,在编排服务时必须满足这些要求。然而,现有的制造服务协作优化方法主要关注服务之间针对功能需求的横向协作,很少考虑纵向协作来覆盖所需的数量。为了解决这一差距,本文提出了一种结合功能和数量协作的二维服务协作方法。首先,提出了一种描述服务的多粒度制造服务建模方法。在此基础上,建立了二维制造服务协同优化模型。在垂直维度上,多个功能等效的服务组成一个服务集群来完成一个子任务;在水平维度上,互补服务集群协作完成整个任务。服务的选择和所选服务的金额分配是模型中的关键问题。为了解决这个问题,设计了一种具有多个局部搜索算子的多目标模因算法。将该算法嵌入竞争机制来动态调整本地搜索算子的选择概率。实验结果表明,与常用算法相比,该算法在收敛性、解质量和综合度量方面具有优势。

摘要：Despite great progress of lithium-sulfur(Li-S)battery performance at the laboratory-level,both key parameters and challenges at cell scales to achieve practical high energy density require high-sulfur-loading cathodes and lean ***,a novel carbon foam integrated by hollow carbon bubble nanoreactors with ultrahigh pore volume of 6.9 cm3·g−1 is meticulously designed for ultrahigh sulfur content up to 96 wt.%.Tailoring polysulfide trapping and ion/electron transport kinetics during the charge-discharge process can be achieved by adjusting the wall thickness of hollow carbon *** a further in-depth understanding of electrochemical reaction mechanism for the cathode is impelled by the in-situ Raman *** a result,the as-prepared cathode delivers high specific capacitances of 1,269 and 695 mAh·g−1 at 0.1 and 5 C,***,Li-S pouch cells with high areal sulfur loading of 6.9 mg·cm−2 yield exceptional practical energy density of 382 Wh·kg−1 under lean electrolyte of 3.5µL·mg−1,which demonstrates the great potential for realistic high-energy Li-S batteries.

摘要：氮杂环丙烷是重要的有机结构骨架,可作为关键中间体构建结构多样的含氮化合物。有机电化学已经成为一种环境友好和可持续的有机合成工具,可以在不依赖外部氧化还原试剂的情况下实现各种化学转化。本文综述了电化学烯烃氮杂环丙烷化合成氮杂环丙烷的最新研究进展,重点介绍了这些转化的反应设计、活性中间体以及发展前景。