Design and implementation of dual-mode configurable memory architecture for CNN accelerator

With the rapid development of deep learning algorithms,the computational complexity and functional diversity are increasing rapidly.However,the gap between high computational density and insufficient memory bandwidth under the traditional von Neumann architecture is getting worse.Analyzing the algorithmic characteristics of convolutional neural network(CNN),it is found that the access characteristics of convolution(CONV)and fully connected(FC)operations are very different.Based on this feature,a dual-mode reronfigurable distributed memory architecture for CNN accelerator is designed.It can be configured in Bank mode or first input first output(FIFO)mode to accommodate the access needs of different operations.At the same time,a programmable memory control unit is designed,which can effectively control the dual-mode configurable distributed memory architecture by using customized special accessing instructions and reduce the data accessing delay.The proposed architecture is verified and tested by parallel implementation of some CNN algorithms.The experimental results show that the peak bandwidth can reach 13.44 GB·s^(-1)at an operating frequency of 120 MHz.This work can achieve 1.40,1.12,2.80 and 4.70 times the peak bandwidth compared with the existing work.

Constructing long-cycling crystalline C_(3)N_(4)-based carbonaceous anodes for sodium-ion battery via N configuration control

Carbon nitrides with two-dimensional layered structures and high theoretical capacities are attractive as anode materials for sodium-ion batteries while their low crystallinity and insufficient structural stability strongly restrict their practical applications.Coupling carbon nitrides with conductive carbon may relieve these issues.However,little is known about the influence of nitrogen(N)configurations on the interactions between carbon and C_(3)N_(4),which is fundamentally critical for guiding the precise design of advanced C_(3)N_(4)-related electrodes.Herein,highly crystalline C_(3)N_(4)(poly(triazine imide),PTI)based all-carbon composites were developed by molten salt strategy.More importantly,the vital role of pyrrolic-N for enhancing charge transfer and boosting Na+storage of C_(3)N_(4)-based composites,which was confirmed by both theoretical and experimental evidence,was spot-highlighted for the first time.By elaborately controlling the salt composition,the composite with high pyrrolic-N and minimized graphitic-N content was obtained.Profiting from the formation of highly crystalline PTI and electrochemically favorable pyrrolic-N configurations,the composite delivered an unusual reverse growth and record-level cycling stability even after 5000 cycles along with high reversible capacity and outstanding full-cell capacity retention.This work broadens the energy storage applications of C_(3)N_(4) and provides new prospects for the design of advanced all-carbon electrodes.

Engineering of geometrical configurations in dual-atom catalysts for electrocatalytic applications

Geometrical configurations play a crucial role in dual-atom catalysts(DACs)for electrocatalytic applications.Significant progress has been made to design DACs electrocatalysts with various geometri-cal configurations,but in-depth understanding the relationship between geometrical configurations and metal-metal interaction mechanisms for designing targeted DACs is still required.In this review,the recent progress in engineering of geometrical configurations of DACs is systematically summarized.Based on the polarity of geometrical configuration,DACs can be classified into two different types that are homonuclear and heteronuclear DACs.Furthermore,with regard to the geometrical configurations of the active sites,homonuclear DACs are identified into adjacent and bridged configurations,and heteronuclear DACs can be classified into adjacent,bridged,and separated configurations.Subsequently,metal-metal interactions in DACs with different geometrical configurations are introduced.Additionally,the applications of DACs in different electrocatalytic reactions are discussed,including the oxygen reduction reaction(ORR),oxygen evolution reaction(OER),hydrogen evolution reaction(HER),and other catalysis.Finally,the future challenges and perspectives for advancements in DACs are high-lighted.This review aims to provide inspiration for the design of highly effcient DACs towards energy relatedapplications.

Review of the Configuration and Transient Stability of Large-scale Renewable Energy Generation through Hybrid DC Transmission

Based on the complementary advantages of Line Commutated Converter(LCC)and Modular Multilevel Converter(MMC)in power grid applications,there are two types of hybrid DC system topologies:one is the parallel connection of LCC converter stations and MMC converter stations,and the other is the series connection of LCC and MMC converter stations within a single station.The hybrid DC transmission system faces broad application prospects and development potential in large-scale clean energy integration across regions and the construction of a new power system dominated by new energy sources in China.This paper first analyzes the system forms and topological characteristics of hybrid DC transmission,introducing the forms and topological characteristics of converter-level hybrid DC transmission systems and system-level hybrid DC transmission systems.Next,it analyzes the operating characteristics of LCC and MMC inverter-level hybrid DC transmission systems,provides insights into the transient stability of hybrid DC transmission systems,and typical fault ride-through control strategies.Finally,it summarizes the networking characteristics of the LCC-MMC series within the converter station hybrid DC transmission system,studies the transient characteristics and fault ridethrough control strategies under different fault types for the LCC-MMC series in the receiving-end converter station,and investigates the transient characteristics and fault ride-through control strategies under different fault types for the LCC-MMC series in the sending-end converter station.

Seismic fragility analysis of three-tower cable-stayed bridges with different connection configurations

Seismic fragility analysis of three-tower cable-stayed bridges with three different structural systems,including rigid system(RS),floating system(FS),and passive energy dissipation system(PEDS),is conducted to study the effects of connection configurations on seismic responses and fragilities.Finite element models of bridges are established using OpenSees.A new ground motion screening method based on the statistical characteristic of the predominant period is proposed to avoid irregular behavior in the selection process of ground motions,and incremental dynamic analysis(IDA)is performed to develop components and systems fragility curves.The effects of damper failure on calculated results for PEDS are examined in terms of seismic response and fragility analysis.The results show that the bridge tower is the most affected component by different structural systems.For RS,the fragility of the middle tower is significantly higher than other components,and the bridge failure starts from the middle tower,exhibiting a characteristic of local failure.For FS and PEDS,the fragility of the edge tower is higher than the middle tower.The system fragility of RS is higher than FS and PEDS.Taking the failure of dampers into account is necessary to obtain reliable seismic capacity of cable-stayed bridges.

Network Configuration Entity Extraction Method Based on Transformer with Multi-Head Attention Mechanism

Nowadays,ensuring thequality of networkserviceshas become increasingly vital.Experts are turning toknowledge graph technology,with a significant emphasis on entity extraction in the identification of device configurations.This research paper presents a novel entity extraction method that leverages a combination of active learning and attention mechanisms.Initially,an improved active learning approach is employed to select the most valuable unlabeled samples,which are subsequently submitted for expert labeling.This approach successfully addresses the problems of isolated points and sample redundancy within the network configuration sample set.Then the labeled samples are utilized to train the model for network configuration entity extraction.Furthermore,the multi-head self-attention of the transformer model is enhanced by introducing the Adaptive Weighting method based on the Laplace mixture distribution.This enhancement enables the transformer model to dynamically adapt its focus to words in various positions,displaying exceptional adaptability to abnormal data and further elevating the accuracy of the proposed model.Through comparisons with Random Sampling(RANDOM),Maximum Normalized Log-Probability(MNLP),Least Confidence(LC),Token Entrop(TE),and Entropy Query by Bagging(EQB),the proposed method,Entropy Query by Bagging and Maximum Influence Active Learning(EQBMIAL),achieves comparable performance with only 40% of the samples on both datasets,while other algorithms require 50% of the samples.Furthermore,the entity extraction algorithm with the Adaptive Weighted Multi-head Attention mechanism(AW-MHA)is compared with BILSTM-CRF,Mutil_Attention-Bilstm-Crf,Deep_Neural_Model_NER and BERT_Transformer,achieving precision rates of 75.98% and 98.32% on the two datasets,respectively.Statistical tests demonstrate the statistical significance and effectiveness of the proposed algorithms in this paper.

A numerical survey of parameters to reach ignition condition for axial compression of a large-sized field reversed configuration

Field reversed configuration(FRC)is widely considered as an ideal target plasma for magnetoinertial fusion.However,its confinement and stability,both proportional to the radius,will deteriorate inevitably during radial compression.Hence,we propose a new fusion approach based on axial compression of a large-sized FRC.The axial compression can be made by plasma jets or plasmoids converging onto the axial ends of the FRC.The parameter space that can reach the ignition condition while preserving the FRC's overall quality is studied using a numerical model based on different FRC confinement scalings.It is found that ignition is possible for a large FRC that can be achieved with the current FRC formation techniques if compression ratio is greater than 50.A more realistic compression is to combine axial with moderate radial compression,which is also presented and calculated in this work.

Configuration Synthesis and Grasping Performance Analysis of Multi-Loop Coupling Capture Mechanism for Launch and Recovery of Torpedo-Shaped Autonomous Underwater Vehicle

Research of capture mechanisms with strong capture adaptability and stable grasp is important to solve the problem of launch and recovery of torpedo-shaped autonomous underwater vehicles(AUVs).A multi-loop coupling capture mechanism with strong adaptability and high retraction rate has been proposed for the launch and recovery of torpedo-shaped AUVs with different morphological features.Firstly,the principle of capturing motion retraction is described based on the appearance characteristics of torpedo-shaped AUVs,and the configuration synthesis of the capture mechanism is carried out using the method of constrained chain synthesis.Secondly,the screw theory is employed to analyze the degree of freedom(DoF)of the capture mechanism.Then,the 3D model of the capture mechanism is established,and the kinematics and dynamics simulations are carried out.Combined with the capture orientation requirements of the capture mechanism,the statics and vibration characteristics analyses are carried out.Furthermore,considering the capture process and the underwater working environment,the motion characteristics and hydraulics characteristics of the capture mechanism are analyzed.Finally,a principle prototype is developed and the torpedo-shaped AUVs capture experiment is completed.The work provides technical reserves for the research and development of AUV capture special equipment.

Green transformation paths of resource-based cities in China from the configuration perspective

Green transformation is an unavoidable choice for resource-based cities(RBCs)that face resource depletion and environmental pollution.Existing research has focused primarily on specific RBCs,making it challenging to apply green transformation strategies universally across cities.The fuzzy set qualitative comparative analysis(fsQCA)is a combination of qualitative and quantitative analyses that can handle multiple concurrent causality problems and determine how different conditions combine into configurations and generate an outcome.Thus,to address this gap,in this study,we established a research framework for green transformation and utilized the fsQCA to examine the configurations of 113 RBCs in China.By incorporating the element of time,this study explored the dynamic evolution of solutions in 2013,2016,and 2019.The main findings indicate that individual elements do not constitute the necessary conditions for improving the green transformation efficiency(GTE),and the systematic combination of multiple conditions is an effective path for realizing the improvement of the GTE in RBCs.Green transformation paths of RBCs exhibit the same destination through different paths.Additionally,the combination of system environment elements and system structure elements is both complementary and alternative.Differences in RBCs have led to various factor combinations and development paths,but there are some similarities in the key elements of the factor combinations at different stages.Economic environment,government support,and technological innovation are key factors that universally enhance the GTE in RBCs.These insights can assist city managers in formulating policies to drive green transformation and contribute to a better theoretical understanding of green transformation paths in RBCs.

Virtually coupled train set control subject to space-time separation:A distributed economic MPC approach with emergency braking configuration

The emerging virtual coupling technology aims to operate multiple train units in a Virtually Coupled Train Set(VCTS)at a minimal but safe distance.To guarantee collision avoidance,the safety distance should be calculated using the state-of-the-art space-time separation principle that separates the Emergency Braking(EB)trajectories of two successive units during the whole EB process.In this case,the minimal safety distance is usually numerically calculated without an analytic formulation.Thus,the constrained VCTS control problem is hard to address with space-time separation,which is still a gap in the existing literature.To solve this problem,we propose a Distributed Economic Model Predictive Control(DEMPC)approach with computation efficiency and theoretical guarantee.Specifically,to alleviate the computation burden,we transform implicit safety constraints into explicitly linear ones,such that the optimal control problem in DEMPC is a quadratic programming problem that can be solved efficiently.For theoretical analysis,sufficient conditions are derived to guarantee the recursive feasibility and stability of DEMPC,employing compatibility constraints,tube techniques and terminal ingredient tuning.Moreover,we extend our approach with globally optimal and distributed online EB configuration methods to shorten the minimal distance among VCTS.Finally,experimental results demonstrate the performance and advantages of the proposed approaches.

Design and implementation of instruction-driven and data-driven self-reconfigurable cell array

The reconfigurable chip,which integrates the advantages of high performance,high flexibility,high parallelism,low power consumption,and low cost,has achieved rapid development and wide application.Generally,the control part and the computing part of algorithm is accelerated based on different reconfigurable architectures,but it is difficult to obtain overall performance improvement.For improving efficiency of reconfigurable structure both for the control part and the computing part,a hybrid of instruction-driven and data-driven self-reconfigurable cell array is proposed.On instruction-driven mode,processing element(PE)works like a reduced instruction set computer(RSIC)machine,which is mainly for the control part of algorithm.On data-driven mode,data is calculated by flowing between the preconfigured PEs,which is mainly for the computing of algorithm.For verifying the efficiency of architecture,some high-efficiency video coding(HEVC)video compression algorithms are implemented on the proposed architecture.The proposed architecture has been implemented on Xilinx FPGA Virtex UltraScale VU440 develop board.The same circuitry is able to run at75 MHz.Compared with the architecture that only supports instruction-driven,the proposed architecture has better calculation efficiency.

基于SPI-RRV指数中国气象干旱及其风险时空演变特征研究

为全面揭示变化环境下我国多维气象干旱特征,耦合气象干旱指数(SPI)和可靠性-回弹性-脆弱性(RRV)指数,提出了一种基于SPI-RRV指数的干旱风险评价方法,定量评价了中国气象干旱及其风险的时空演变特征。结果表明:SPI-RRV指数具有特征稳定和时空可比性强的特点,能够较为准确地评估气象干旱风险时空演变特征;南方平均干旱栅格比、干旱月占比和频次大于北方,湿润区和半湿润区干旱历时短、烈度大,半干旱区和干旱区干旱历时长、烈度相对较小;干旱高风险区转移具有显著年代际变化规律,空间上从西北向西南地区转移。

深部煤储层孔裂隙结构对煤层气赋存的影响-以鄂尔多斯盆地东缘大宁-吉县区块为例

深部煤储层孔隙–裂缝结构对深部煤层气资源潜力评价和勘探开发具有重要意义。选取鄂尔多斯盆地东缘大宁–吉县区块DJ57井本溪组5个煤岩样品为研究对象,在煤岩煤质参数测试的基础上,采用气体吸附法、高压压汞法和微米CT扫描等测试手段,对深部煤储层中的纳米级孔隙-微米级裂缝进行多尺度定量表征,综合评价不同尺度的孔裂隙结构特征。再结合渗透率和甲烷等温吸附试验,探讨了微观孔裂隙对深部煤储层中煤层气的赋存和渗流的影响。研究结果表明:基于多种孔隙表征方法对深部煤储层孔裂隙进行多尺度定量表征,其孔裂隙体积分布类型主要以“U”型为主,呈现出微孔与微裂缝并存双峰态,主要集中在0.3~1.5 nm和>100μm的范围内。其中,微孔(<2 nm)、介孔(2~50 nm)、宏孔(50 nm~10μm)和微裂缝(>10μm)体积平均分别占总孔裂隙体积的80.18%,6.70%,1.65%和11.47%。随着微孔发育而吸附气量呈增大的趋势,微孔可以提供大量吸附点位,为深部煤层气的吸附和赋存提供场所。随着微裂缝发育而游离气量呈增大的趋势,微裂缝可以提供大量储集空间,为深部煤层气的富集提供空间条件。此外,微裂缝在三维空间中相互连通,形成网状结构,连通性强。随着微裂缝越发育,煤储层渗透率越大,微裂缝增强了煤层气的渗流能力。纳米级孔隙和微米级裂隙发育特征分别控制着深部煤层气吸附能力和开发潜力。

水资源-能源-粮食-生态系统耦合协调及驱动力分析

为加深对我国水资源、能源、粮食、生态系统协同演变趋势的认识,构建水资源-能源-粮食-生态多维系统指标体系,运用耦合协调度模型对我国2005—2020年水资源-能源-粮食-生态系统耦合协调度进行评价,并采用多因素归因分析法进行驱动力分析。结果表明:我国水资源-能源-粮食-生态系统耦合协调度从2005年的0.55增长到2020年的0.84,各地区耦合协调度从勉强协调发展水平过渡到中级协调发展水平,各子系统对耦合协调度上升的驱动分别经历了由粮食子系统到生态子系统再到水资源子系统主导的过程;能源子系统的贡献率虽然比较小,但是未来可能是各地区提升水资源-能源-粮食-生态系统多维系统协调发展水平的突破口。

基于DEMATEL-ISM的水利工程EPC项目价值增值机理研究

采用设计-采购-施工(engineering-procurement-construction,EPC)总承包模式的水利工程项目日益增多,研究其价值增值机理有利于总承包方做好项目的价值管理。从水利工程EPC项目的价值链分析入手,基于文献和专家访谈得到23个价值增值影响因素。采用问卷调查方式得到各因素间的影响程度评分,用决策实验和评估实验(decision-making trial and evaluation laboratory,DEMATEL)法计算了各因素的中心度和原因度,结合解释结构模型(interpretative structural modeling,ISM)对数据进行分析,将价值增值因素划分为6类,构建了水利工程EPC项目价值增值的递阶模型,探究和揭示了项目价值增值的路径和机理。研究结果对水利工程EPC总承包方制订管理策略有一定启示作用。

基于UPLC-QTOF-MS解析苦荞黄酮组成及药理学网络分析

为优化苦荞黄酮提取工艺并明确其具体成分结构和含量及药理作用,本研究以四川省凉山彝族自治州甘洛县优质苦荞为研究对象,对超声提取苦荞黄酮工艺进行了优化,采用超高效液相色谱-四级杆-飞行时间质谱(UPLC-QTOF-MS)对主要黄酮成分进行结构鉴定,并采用超高效液相色谱-三重四级杆质谱联用仪对鉴定到的黄酮进行准确定量,同时利用中药系统药理学(TCMSP)系统分析其网络药理学特性。结果表明,超声辅助乙醇提取苦荞黄酮的最优条件为乙醇浓度90%(V/V)、超声功率140 W、超声温度70℃、超声时间60 min、提取溶剂与苦荞原料比50∶1(mL·g^(-1)),该条件下,提取液中共鉴定到黄酮16种,其中芦丁、表儿茶素和儿茶素含量分别达到115.81、72.99和23.08μg·mL^(-1)。这16种黄酮在防治心血管疾病、慢性炎症性疾病上具有重要作用。本研究可为苦荞精深加工综合利用尤其是功能性苦荞产品开发利用奠定基础。

基于“心-脉-神”互言探讨肝心同治冠心病合并抑郁(双心疾病)的理论研究

双心疾病作为“血脉之心”与“神明之心”同病的复合疾病,已经成为现代社会影响人类健康的主流问题之一,其中的冠心病合并抑郁属于临床中常见的发病类型。冠心病是一种严重威胁健康的心血管疾病,焦虑和抑郁是其不良预后的潜在危险因素。冠心病合并抑郁属中医“胸痹”“郁证”范畴,“心藏脉,脉舍神”,心脏对生命活动具有重要的控制和调节作用,“从心论治”使机体心主神明、心主血脉功能正常发挥。心血管疾病与情志病是密不可分的,胸痹患者心脉不畅,肝失疏泄,痰瘀互结,气机郁滞加重,由此引发情志的改变。双心疾病与中医的“心主血脉”“心主神明”“肝主藏血”“肝主疏泄”功能失常密切相关,“心”与“肝”的状态失衡是导致疾病进展的主要因素,故“肝心同治”是调节关键病机-肝心失调的必由之法,以疏肝解郁、调气和血为主。基于对“从心论治”“心-脉-神”与“肝心同治”的理论探讨,结合现代科学技术,旨在为防治冠心病合并抑郁提供新的思路与机制预测。

基于混合式教学的“教-学-评”一体化实验教学体系探索

实验能力是医学专业研究生的重要科研能力,实验能力的培养与实验教学的方法与质量密切相关。文章针对传统实验教学受到时间和空间限制且评价指标较为单一的现状,提出构建基于线上自主学习和现场实践相结合的“教-学-评”一体化实验教学体系,建立覆盖课前线上自主学习、课堂现场实践、课后作业提交与考核教学全过程的评价体系,以评促学,以评促教,通过评价反馈激发学生的学习热情,改进教学内容和方法,不断探索与完善实验教学体系,提高实验教学质量,以期为培养具有实践能力与创新精神的医学科研人才发挥积极作用。

制造企业智能化转型绩效影响因素的组态效应——基于“动机-能力-机会”(AMO)理论分析框架

制造企业智能化转型是多种要素协同作用的结果,而如何提升转型绩效是当前中国制造企业面临的主要难题。但学界针对此议题的深入研究尚较为缺乏。对此,首先对制造企业智能化转型概念作出定义并分析其内涵要素;其次,鉴于组织、团队或个体的绩效水平会受到其能力(ability)、动机(motivation)和机会(opportunity)等多方面因素的影响,基于“动机-能力-机会”(AMO)理论框架构建制造企业智能化转型绩效影响因素的组态模型;最后,采用面向全国制造企业的103份有效调查问卷数据,并结合典型案例,运用模糊集定性比较分析方法(fsQCA)分析企业转型能力(企业动态能力、企业知识禀赋)、转型动机(内在转型需求、外部竞争压力)以及转型机会(政府政策支持、数字基础设施)对于样本企业智能化转型绩效的组态效应。结果表明:能力、动机和机会均无法单独构成制造企业高智能化转型绩效的必要条件;存在8种会导致企业高智能化转型绩效的条件组态并可归纳为能力主导机会支持型、动机主导能力支持型、动机主导机会支持型、机会主导能力支持型和能力、动机、机会兼备型五大组态类型;有1种条件组态会导致企业非高智能化转型绩效,可归纳为能力、动机、机会兼不具备型。由此得到启示:制造企业在向智能制造转型升级的过程中须重视对自身能力、动机和机会三者情况进行综合性考量,结合实际选择合理的转型升级路径。

基于HS-GC-IMS和HS-SPME-GC-MS的蛋白酶对豆粕挥发性风味的影响分析

为研究不同蛋白酶酶解对豆粕挥发性风味成分的影响,选用4种蛋白酶(碱性蛋白酶、中性蛋白酶、木瓜蛋白酶、风味蛋白酶)对豆粕进行酶解,采用顶空-气相色谱-离子迁移谱(Headspace-gas chromatography-ion mobility spectroscopy,HS-GC-IMS)和顶空固相微萃取-气相色谱-质谱(Headspace solid phase microextraction-gas chromatography-mass spectrometry,HS-SPME-GC-MS)联用技术分析不同豆粕酶解物(Soybean meal hydrolysates,SMH)的挥发性风味成分,并结合主成分分析(Principal component analysis,PCA)、热图聚类和正交偏最小二乘判别法(Orthogonal partial least squares-discriminant analysis,OPLS-DA)对不同SMH进行分析。结果表明:碱性蛋白酶、中性蛋白酶、木瓜蛋白酶和风味蛋白酶酶解豆粕的挥发性风味成分存在较大差异。HS-GC-IMS鉴定出84种挥发性成分,筛选得到33种差异风味物质,发现酶解后酮类物质显著降低而醛类、醇类和酯类物质含量明显增加。PCA结果表明不同SMH之间的风味存在显著差异。最终通过OPLS-DA筛选出贡献较大的挥发性化合物,同时构建出可靠的用以鉴别SMH的模型。HS-SPME-GC-MS检测出103种差异风味物质,可用于区分不同SMH,被检出的挥发性组分中醛类、醇类和酮类等化合物为SMH风味的形成做出主要贡献,明晰了部分风味化合物形成的原因。PCA和聚类热图结果表明不同蛋白酶酶解对豆粕的挥发性风味物质的种类和含量有显著影响,其中,风味蛋白酶和木瓜蛋白酶对豆粕的风味改善最为显著。