期刊文献+
共找到2,745篇文章
< 1 2 138 >
每页显示 20 50 100
基于Matrix Core的高性能多维FFT设计与优化 被引量:2
1
作者 陆璐 祝松祥 +2 位作者 田卿燕 林海山 郭逸劼 《华南理工大学学报(自然科学版)》 北大核心 2025年第3期20-30,共11页
快速傅里叶变换(FFT)算法广泛应用于科学计算等领域。为了充分挖掘图形处理器(GPU)的计算能力并进一步提高FFT的计算效率,该文针对矩阵形式的Stockham FFT,提出了一种基于Matrix Core的高性能多维FFT计算方案。在计算优化方面,该方案利... 快速傅里叶变换(FFT)算法广泛应用于科学计算等领域。为了充分挖掘图形处理器(GPU)的计算能力并进一步提高FFT的计算效率,该文针对矩阵形式的Stockham FFT,提出了一种基于Matrix Core的高性能多维FFT计算方案。在计算优化方面,该方案利用Matrix Core加速FFT计算中的矩阵乘运算,同时通过编译器内部指令完成小粒度的矩阵乘加,使得Matrix Core支持更多尺寸的FFT计算。在内存优化方面,该方案使用2层迭代策略,以充分利用共享内存,减少与全局内存的数据交换;根据Matrix Core的矩阵数据在各个线程寄存器中的分布规律,直接在寄存器上完成FFT计算中大量存在的矩阵逐元素乘操作;通过对共享内存中的数据进行重排来缓解存储体冲突,并采用双缓冲策略缓解访存瓶颈。该文还提出了高效的矩阵转置策略,以加速多维FFT计算。在AMD MI250 GPU平台上将该方案与GPU上主流的高性能FFT计算库rocFFT和VkFFT进行了比较实验,结果表明:该方案在AMD MI250上的1维、2维和3维FFT平均计算效率均优于rocFFT和VkFFT,3维FFT的平均计算效率为rocFFT的1.5倍,为VkFFT的2.0倍,具有较好的性能提升;mcFFT的计算精度与rocFFT和VkFFT保持在相同水平。 展开更多
关键词 图形处理器 matrix Core 快速傅里叶变换 矩阵乘法
在线阅读 下载PDF
基于Matrix Core的小尺寸批量矩阵乘法设计与优化 被引量:1
2
作者 陆璐 赵容 +1 位作者 梁志宏 索思亮 《华南理工大学学报(自然科学版)》 北大核心 2025年第9期48-58,共11页
通用矩阵乘法(GEMM)是线性代数中最重要的运算,来自不同科学领域的许多应用程序都将其关键部分转换为使用GEMM的形式。GEMM广泛应用于大模型、机器学习、科学计算和信号处理等领域。特别是半精度的批处理GEMM(即FP16)一直是许多深度学... 通用矩阵乘法(GEMM)是线性代数中最重要的运算,来自不同科学领域的许多应用程序都将其关键部分转换为使用GEMM的形式。GEMM广泛应用于大模型、机器学习、科学计算和信号处理等领域。特别是半精度的批处理GEMM(即FP16)一直是许多深度学习框架的核心操作。目前AMD GPU上半精度批处理GEMM的访存和计算利用率不足,急需优化。为此,该文提出了一种半精度批处理GEMM(HGEMM)的图形处理器(GPU)优化方案。分块策略方面,根据输入矩阵块大小为线程分配相同的访存量和计算量,同时线程计算多个矩阵乘法,以提高计算单元的利用率。访存优化方面,以多读数据为代价,为每个线程分配相同访存量以便于编译器优化,保证访存和计算时间相互掩盖。对于矩阵尺寸小于16的极小尺寸批处理HGEMM,该文利用4×4×4的Matrix Core及其对应的分块方案,在提升访存性能的同时减少Matrix Core计算资源的浪费,并提供是否使用共享内存的选项来达到最高性能。在AMD GPU MI210平台上,将该方案与rocBLAS的2个算子进行性能对比,结果表明:该方案在AMD GPU MI210上的平均性能为rocBLASHGEMMBatched的4.14倍,rocBLASGEMMExBatched的4.96倍;对于极小尺寸批处理HGEMM,平均性能为rocBLASHGEMMBatched的18.60倍,rocBLASGEMMExBatched的14.02倍。 展开更多
关键词 图形处理器 matrix Core 矩阵乘法 访存优化
在线阅读 下载PDF
Marginal optimization method to improve the inconsistent comparison matrix in the analytic hierarchy process 被引量:12
3
作者 Shihui Wu Jiang Xie +3 位作者 Xiaodong Liu Bo He Minxiang Yang Zhengxin Li 《Journal of Systems Engineering and Electronics》 SCIE EI CSCD 2017年第6期1141-1151,共11页
To improve the inconsistency in the analytic hierarchy process(AHP), a new method based on marginal optimization theory is proposed. During the improving process, this paper regards the reduction of consistency ratio(... To improve the inconsistency in the analytic hierarchy process(AHP), a new method based on marginal optimization theory is proposed. During the improving process, this paper regards the reduction of consistency ratio(CR) as benefit, and the maximum modification compared to the original pairwise comparison matrix(PCM) as cost, then the improvement of consistency is transformed to a benefit/cost analysis problem. According to the maximal marginal effect principle, the elements of PCM are modified by a fixed increment(or decrement) step by step till the consistency ratio becomes acceptable, which can ensure minimum adjustment to the original PCM so that the decision makers’ judgment is preserved as much as possible. The correctness of the proposed method is proved mathematically by theorem. Firstly, the marginal benefit/cost ratio is calculated for each single element of the PCM when it has been modified by a fixed increment(or decrement).Then, modification to the element with the maximum marginal benefit/cost ratio is accepted. Next, the marginal benefit/cost ratio is calculated again upon the revised matrix, and followed by choosing the modification to the element with the maximum marginal benefit/cost ratio. The process of calculating marginal effect and choosing the best modified element is repeated for each revised matrix till acceptable consistency is reached, i.e., CR<0.1. Finally,illustrative examples show the proposed method is more effective and better in preserving the original comparison information than existing methods. 展开更多
关键词 analytic hierarchy process(AHP) comparison matrix CONSISTENCY marginal optimization
在线阅读 下载PDF
Optimization of cold-sprayed AA2024/Al_2O_3 metal matrix composites via friction stir processing: Effect of rotation speeds 被引量:4
4
作者 Kang Yang Wenya Li +2 位作者 Chunjie Huang Xiawei Yang Yaxin Xu 《Journal of Materials Science & Technology》 SCIE EI CAS CSCD 2018年第11期2167-2177,共11页
In this study, friction stir processing(FSP) was employed to modify cold-sprayed(CSed) AA2024/Al2 O3 metal matrix composites(MMCs). Three different rotation speeds with a constant traverse speed were used for FS... In this study, friction stir processing(FSP) was employed to modify cold-sprayed(CSed) AA2024/Al2 O3 metal matrix composites(MMCs). Three different rotation speeds with a constant traverse speed were used for FSP. Microstructural analysis of the FSPed specimens reveals significant Al2 O3 particle refinement and improved particle distribution over the as-sprayed deposits. After FSP, a microstructural and mechanical gradient MMC through the thickness direction was obtained. Therefore, a hybrid technique combining these two solid-state processes, i.e. CS and FSP, was proposed to produce functionally gradient deposits. The Guinier-Preston-Bagaryatskii zone was dissolved during FSP, while the amounts at different rotation speeds were approximately the same, which is possibly due to the excellent thermal conductivity of the used Cu substrate. Mechanical property tests confirm that FSP can effectively improve the tensile performance and Vickers hardness of CSed AA2024/Al2 O3 MMCs. The properties can be further enhanced with a larger rotation speed with a maximum increase of 25.9% in ultimate tensile strength and27.4% in elongation at 1500 rpm. Friction tests show that FSP decreases the wear resistance of CSed MMCs deposits due to the breakup of Al2 O3 particles. The average values and fluctuations of friction coefficients at different rotation speeds vary significantly. 展开更多
关键词 Cold spray Friction stir processing Metal matrix composite Functionally gradient deposit Mechanical property TRIBOLOGY
原文传递
Processing,Microstructures,and Mechanical Properties of Magnesium Matrix Composites:A Review 被引量:16
5
作者 Liqing Chen Yantao Yao 《Acta Metallurgica Sinica(English Letters)》 SCIE EI CAS CSCD 2014年第5期762-774,共13页
In the last two decades, light-weight magnesium matrix composites have been the hot issue of material field due to their excellent mechanical and physical properties, e.g., high-specific strength and modulus, good wea... In the last two decades, light-weight magnesium matrix composites have been the hot issue of material field due to their excellent mechanical and physical properties, e.g., high-specific strength and modulus, good wear resistance, and damping capacity. As compared with aluminum matrix composites, magnesium matrix composites have merit in their specific weight and have wide applications in aerospace and aeronautical fields. Generally, the processing techniques for magnesium matrix composites can be categorized as conventional and special processing routes. In recent years, as a special processing route, metal melt infiltration into porous ceramic preform featured by its low cost and availability of high-volume fraction of reinforced ceramics have been receiving much attention. Thus, in this review, one emphasis was put on the description of this processing technique in association with the means to obtain good wettability, the prerequisite for this kind of processing method. Based on the recognized fact that there exist clean interface and bonding ability between ceramics and matrix metal, in-situ reaction synthesis is usually utilized to fabricate magnesium matrix composites. Therefore, the interfacial feature was also reviewed for the in-situ reaction synthesis. Characterizations of microstructures and various mechanical-physical properties were finally summarized for magnesium matrix composites including tensile response, wear resistance, creep behavior, and damping capacity. 展开更多
关键词 Magnesium matrix composites processING Microstructure Mechanical properties WETTABILITY INTERFACE
原文传递
Preparation of Aluminum Matrix Composite by Pressureless Infiltration Process
6
作者 李崇俊 马伯信 王抗利 《Rare Metals》 SCIE EI CAS CSCD 1998年第1期6-10,共5页
he Al_2O_3 particle reinforced aluminum matrix composite was prepared by using a new pressureless infiltration process. The microstructure of (Al_2O_3)p/Al was analyzed. The tension and the thermal conductivity of the... he Al_2O_3 particle reinforced aluminum matrix composite was prepared by using a new pressureless infiltration process. The microstructure of (Al_2O_3)p/Al was analyzed. The tension and the thermal conductivity of the composite were studied as well. 展开更多
关键词 Aluminum matrix composite Pressureless infiltration process
在线阅读 下载PDF
Hot deformation and processing maps of titanium matrix composite
7
作者 王蕊宁 奚正平 +1 位作者 赵永庆 戚运莲 《中国有色金属学会会刊:英文版》 CSCD 2007年第A01期541-545,共5页
The hot deformation characteristics of TiC particles reinforced titanium matrix composite were studied in the temperature range from 900 ℃ to 1 150 ℃ and in the strain rate range of 10-3-10 s-1 by compression tests ... The hot deformation characteristics of TiC particles reinforced titanium matrix composite were studied in the temperature range from 900 ℃ to 1 150 ℃ and in the strain rate range of 10-3-10 s-1 by compression tests with Gleeble1500 simulator system. The flow behavior was described by the hyperbolic sine constitutive equation,and an average activation energy of 436.72 kJ/mol was calculated. The processing maps were calculated and analyzed according to the dynamic materials model. The maps show domains in some combinations of temperatures and strain rates and these domains are correlated with specific microstructural processes occurring during hot deformation by metallographic investigations and kinetic analysis. At the low strain rate domain occurs in the temperature range of 900-960 ℃ and strain rate range of 0.001-0.03 s-1 superplasticity and dynamic recrystallization were observed. At a high strain rate domain occurs in the temperature range of 980-1 120 ℃ and strain rate range of 0.1-10 s-1 the β phase undergoes dynamic recrystallization. Also,at a strain rate range of 0.1-10 s-1 and the temperature range of 900-930 ℃,the material exhibits flow localization. 展开更多
关键词 矩阵 化合物 热成型技术
在线阅读 下载PDF
Microstructural Characterization and Tensile Behavior of Rutile(TiO_2)-Reinforced AA6063 Aluminum Matrix Composites Prepared by Friction Stir Processing
8
作者 Sahayam Joyson Abraham Isaac Dinaharan +1 位作者 Jebaraj David Raja Selvam Esther Titilayo Akinlabi 《Acta Metallurgica Sinica(English Letters)》 SCIE EI CAS CSCD 2019年第1期52-62,共11页
Rutile(TiO_2) particle-reinforced aluminum matrix composites were prepared by friction stir processing. The microstructure was studied using conventional and advanced characterization techniques. TiO_2 particles were ... Rutile(TiO_2) particle-reinforced aluminum matrix composites were prepared by friction stir processing. The microstructure was studied using conventional and advanced characterization techniques. TiO_2 particles were found to be dispersed uniformly in the composite. Clusters of TiO_2 particles were observed at a higher particle content of 18 vol%. The interface between the TiO_2 particle and the aluminum matrix was characterized by the absence of pores and reactive layer.Sub-grain boundaries, ultra-fine grains and dislocation density were observed in the composites. TiO_2 particles improved the mechanical properties of the composites. However, a drop in tensile strength was observed at a higher particle content due to cluster formation. All the prepared composites exhibited ductile mode of fracture. 展开更多
关键词 Aluminum matrix composites FRICTION STIR processing RUTILE Microstructure TENSILE strength
原文传递
Tensile Strength and Electrical Conductivity of Carbon Nanotube Reinforced Aluminum Matrix Composites Fabricated by Powder Metallurgy Combined with Friction Stir Processing 被引量:23
9
作者 Z.Y.Liu B.L.Xiao +1 位作者 W.G.Wang Z.Y.Ma 《Journal of Materials Science & Technology》 SCIE EI CAS CSCD 2014年第7期649-655,共7页
A route combining powder metallurgy and subsequent friction stir processing was utilized to fabricate carbon nanotube (CNT) reinforced AI (CNT/AI) and 6061AI (CNT/6061AI) composites. Microstructural observations... A route combining powder metallurgy and subsequent friction stir processing was utilized to fabricate carbon nanotube (CNT) reinforced AI (CNT/AI) and 6061AI (CNT/6061AI) composites. Microstructural observations indicated that CNTs were uniformly dispersed in the matrix in both CNT/AI and CNT/6061AI composites. Mg and Si elements tended to segregate at CNT-AI interfaces in the CNT/6061AI composite during artificial aging treatment. The tensile properties of both the AI and 6061AI were increased by CNT incorporation. The electrical conductivity of CNT/AI was decreased by CNT addition, while CNT/6061AI exhibited an increase in electrical conductivity due to the Mg and Si segregation. 展开更多
关键词 Carbon nanotubes Metal matrix composites Mechanical properties Electrical properties Friction stir processing
原文传递
Microstructure,mechanical properties,and wear resistance of VCp-reinforced Fe-matrix composites treated by Q&P process 被引量:2
10
作者 Ping-hu Chen Yi-bo Li +3 位作者 Rui-qing Li Ri-peng Jiang Song-sheng Zeng Xiao-qian Li 《International Journal of Minerals,Metallurgy and Materials》 SCIE EI CAS CSCD 2018年第9期1060-1069,共10页
A quenching and partitioning(Q&P) process was applied to vanadium carbide particle(VCp)-reinforced Fe-matrix composites(VC-Fe-MCs) to obtain a multiphase microstructure comprising VC, V8 C7, M3 C, α-Fe, and γ-Fe... A quenching and partitioning(Q&P) process was applied to vanadium carbide particle(VCp)-reinforced Fe-matrix composites(VC-Fe-MCs) to obtain a multiphase microstructure comprising VC, V8 C7, M3 C, α-Fe, and γ-Fe. The effects of the austenitizing temperature and the quenching temperature on the microstructure, mechanical properties, and wear resistance of the VC-Fe-MCs were studied. The results show that the size of the carbide became coarse and that the shape of some particles began to transform from diffused graininess into a chrysanthemum-shaped structure with increasing austenitizing temperature. The microhardness decreased with increasing austenitizing temperature but substantially increased after wear testing compared with the microhardness before wear testing; the microhardness values improved by 20.0% ± 2.5%. Retained austenite enhanced the impact toughness and promoted the transformation-induced plasticity(TRIP) effect to improve wear resistance under certain load conditions. 展开更多
关键词 vanadium CARBIDE Fe-matrix composites QUENCHING and partitioning process transformation-induced plasticity effect MICROHARDNESS impact toughness wear resistance
在线阅读 下载PDF
Semisolid-rolling and annealing process of woven carbon fibers reinforced Al-matrix composites 被引量:8
11
作者 Junjia Zhang Shichao Liu +3 位作者 Yiping Lu Li Jiang Yubo Zhang Tingju Li 《Journal of Materials Science & Technology》 SCIE EI CAS CSCD 2017年第7期623-629,共7页
Semisolid-rolling method was successfully developed to prepare the Ni-coated woven carbon fibers reinforced Al-matrix composite. Due to the appropriate matrix flowability and rolling pressure, the Al-matrix could infi... Semisolid-rolling method was successfully developed to prepare the Ni-coated woven carbon fibers reinforced Al-matrix composite. Due to the appropriate matrix flowability and rolling pressure, the Al-matrix could infiltrate into the woven fibers sufficiently and attach to the reinforcements closely forming a smooth interface. The rolling speed of 4 rad/min offered a subtle equilibrium between the heat transfer and the material deformation. The covering matrix should be controlled at semisolid state to provide a better infiltration behavior and a protective effect on the carbon fibers. With the addition of fibers, an improvement for more than 25% was obtained in the bending strength of the materials. Furthermore, the woven carbon fibers could strengthen the composite in multiple directions, rather than only along the fiber longitudinal directions. The annealing process promoted the Ni coating to react with and to diffuse into the matrix, resulted in an obvious increase of the bending strength. 展开更多
关键词 Metal-matrix composites Woven carbon fiber Semisolid-rolling process Annealing Mechanical property
原文传递
Structures and Properties of Iron Matrix Composites with Tungsten Carbide Particle by EPC-V Process 被引量:5
12
作者 Xu Daqing Luo Jirong Huang Naiyu 《Journal of Iron and Steel Research International》 SCIE EI CAS CSCD 1999年第2期29-32,共4页
In this paper the Expendable Pattern Casting with dry sand Vacuum(EPC-V) process is used to manufacture iron matrix composites with tungsten carbide particle.Microstructures of the composites layers were analyzed.The ... In this paper the Expendable Pattern Casting with dry sand Vacuum(EPC-V) process is used to manufacture iron matrix composites with tungsten carbide particle.Microstructures of the composites layers were analyzed.The abrasive wear resistance of the composites layers were tested and compared with that of high chromium cast iron.The results show that the iron matrix composites with tungsten carbide particle have high hardness.The abrasive wear resistance of composites with tungsten carbide particle is higher than that of high chromium cast iron.The properties of the matrix materials have been improved remarkably. 展开更多
关键词 ABSTRACT:In this paper the Expendable Pattern Casting with dry sand Vacuum (EPC V) process is used to manufacture iron matrix COMPOSITES with TUNGSTEN CARBIDE particle. Microstructures of the COMPOSITES layers were analyzed. The ABRASIVE wear
在线阅读 下载PDF
面向GPU的稀疏对角矩阵自适应SpMV优化方法
13
作者 王宇华 何俊飞 +2 位作者 张宇琪 兰海燕 曹林琳 《计算机工程》 北大核心 2026年第3期332-345,共14页
稀疏矩阵向量乘(SpMV)是稀疏线性系统的计算核心和瓶颈,其运算效率会影响迭代求解器的整体性能,其优化研究一直是科学计算和工程应用领域中的研究热点之一。偏微分方程的离散化会产生稀疏对角矩阵,由于其多样的非零元分布,导致没有一种... 稀疏矩阵向量乘(SpMV)是稀疏线性系统的计算核心和瓶颈,其运算效率会影响迭代求解器的整体性能,其优化研究一直是科学计算和工程应用领域中的研究热点之一。偏微分方程的离散化会产生稀疏对角矩阵,由于其多样的非零元分布,导致没有一种方法能够在所有矩阵中取得最优时间性能。针对上述问题,提出一种面向图形处理单元(GPU)的稀疏对角矩阵自适应SpMV优化方法AST(Adaptive SpMV Tuning)。该方法通过设计特征空间,构建特征提取器,提取矩阵结构精细特征,通过深入分析特征和SpMV方法的相关性,建立可扩展的候选方法集合,形成特征和最优方法的映射关系,构建性能预测工具,实现矩阵最优方法的高效预测。实验结果表明,AST能够取得85.8%的预测准确率,平均时间性能损失为0.09,相比于DIA(Diagonal)、HDIA(Hacked DIA)、HDC(Hybrid of DIA and Compressed Sparse Row)、DIA-Adaptive和DRM(Divide-Rearrange and Merge),能够获得平均20.19、1.86、3.06、3.72和1.53倍的内核运行时间加速和1.05、1.28、12.45、1.94和0.97倍的浮点运算性能加速。 展开更多
关键词 稀疏矩阵向量乘 稀疏对角矩阵 图形处理单元 自适应优化方法 矩阵结构特征
在线阅读 下载PDF
食物基质对类胡萝卜素生物可及性的影响研究进展
14
作者 顾叶 刘回民 《粮食与油脂》 北大核心 2026年第3期22-27,共6页
综述了单一食物基质(脂质、蛋白质、碳水化合物等)及复杂食物基质对类胡萝卜素生物可及性的影响,深入探讨了其影响机制,包括物理屏障作用、相互作用等;同时总结了食品加工方式对类胡萝卜素生物可及性的影响,以期为功能性食品开发、营养... 综述了单一食物基质(脂质、蛋白质、碳水化合物等)及复杂食物基质对类胡萝卜素生物可及性的影响,深入探讨了其影响机制,包括物理屏障作用、相互作用等;同时总结了食品加工方式对类胡萝卜素生物可及性的影响,以期为功能性食品开发、营养强化配方设计及类胡萝卜素高效利用提供理论支持。 展开更多
关键词 类胡萝卜素 生物可及性 食物基质 食品加工
在线阅读 下载PDF
基于Matrix VB的数字图像处理教学演示软件 被引量:3
15
作者 余鹏飞 郭磊 《信息技术》 2003年第12期93-94,97,共3页
数字图像处理教学演示软件对于数字图像处理课程的建设具有重要的意义,如果使用MATLAB或者其它软件开发工具来进行开发,软件的开发周期长,发布较麻烦。在介绍MatrixVB的基础上,给出了利用MatrixVB在VisualBasic下进行数字图像处理教学... 数字图像处理教学演示软件对于数字图像处理课程的建设具有重要的意义,如果使用MATLAB或者其它软件开发工具来进行开发,软件的开发周期长,发布较麻烦。在介绍MatrixVB的基础上,给出了利用MatrixVB在VisualBasic下进行数字图像处理教学演示软件的方法和步骤。 展开更多
关键词 数字图像处理 MATLAB VB matrix VB
在线阅读 下载PDF
一种多精度可重构张量计算单元的设计
16
作者 胡湘宏 梁克龙 +4 位作者 尹飞跃 冯兆樟 林元妙 蔡述庭 熊晓明 《集成电路与嵌入式系统》 2026年第3期81-89,共9页
随着人工智能与深度学习应用的快速发展,张量计算对高能效、多精度计算硬件加速器提出了迫切需求。传统通用处理器在处理大规模矩阵乘法运算时存在能效瓶颈,而现有专用加速器往往难以灵活支持多种数据精度与混合计算模式。文中基于可重... 随着人工智能与深度学习应用的快速发展,张量计算对高能效、多精度计算硬件加速器提出了迫切需求。传统通用处理器在处理大规模矩阵乘法运算时存在能效瓶颈,而现有专用加速器往往难以灵活支持多种数据精度与混合计算模式。文中基于可重构架构设计了一款支持多精度与混合精度的张量处理单元,支持INT4、INT8、FP16、BF16、FP32五种数据精度及FP16+FP32、BF16+FP32两种混合精度模式,可高效完成3种不同维度(m16n16k16、m32n8k16、m8n32k16)的矩阵乘加运算。通过可重构计算阵列、动态数据流控制、多模式缓存设计及统一的浮点处理单元,实现了硬件复用率与计算效率的显著提升。在VCU118 FPGA平台上综合频率达251.13 MHz,算力最高达257.16 GOPS/GFLOPS(INT4/INT8/FP16/BF16)和64.29 GFLOPS(FP32)。该设计可广泛应用于深度学习推理、自动驾驶、医疗影像等对计算能效和灵活性要求较高的场景。 展开更多
关键词 张量处理单元 多精度计算 可重构架构 矩阵乘法 硬件复用
在线阅读 下载PDF
基于数据挖掘算法的配电网全过程造价控制方法研究
17
作者 陈付雷 李建青 施晓敏 《电子设计工程》 2026年第5期106-110,116,共6页
从全过程的角度分析,影响配电网造价的因素较多,且影响程度不同。为此,提出基于数据挖掘算法的配电网全过程造价控制方法。该方法分别从设计阶段、施工过程、材料设备采购三个方面,分析配电网全过程造价的影响因素构成,采用两两比较的... 从全过程的角度分析,影响配电网造价的因素较多,且影响程度不同。为此,提出基于数据挖掘算法的配电网全过程造价控制方法。该方法分别从设计阶段、施工过程、材料设备采购三个方面,分析配电网全过程造价的影响因素构成,采用两两比较的方式构建配电网全过程造价影响因素的判断矩阵,对其进行归一化处理,结合最大特征根参量为影响因素赋权。在控制阶段,采用层次分析法构建分层控制目标体系,包括总体控制目标、子控制目标、控制阶段目标与控制指标。将敏感系数趋近于0的影响因素状态作为最终控制结果,以降低其对造价的扰动。实验结果显示,设计控制方法下的单位投资成本明显低于对照组。 展开更多
关键词 数据挖掘算法 配电网全过程 造价控制 判断矩阵 归一化处理 最大特征根 敏感系数
在线阅读 下载PDF
基于图像处理的封印Data Matrix码验证方法 被引量:3
18
作者 李波 曹敏 +4 位作者 李仕林 杨明 林聪 姚鸿泰 王先培 《武汉大学学报(工学版)》 CAS CSCD 北大核心 2019年第11期1020-1027,共8页
在对电能计量装置上的Data Matrix码进行识别过程中,拍摄角度、光照等因素会给识别过程带来困难.针对这一问题,提出一种基于局部阈值与Otsu算法的Data Matrix码识别算法.该算法首先将彩色图像进行灰度化处理,并使用中值滤波算法滤除噪声... 在对电能计量装置上的Data Matrix码进行识别过程中,拍摄角度、光照等因素会给识别过程带来困难.针对这一问题,提出一种基于局部阈值与Otsu算法的Data Matrix码识别算法.该算法首先将彩色图像进行灰度化处理,并使用中值滤波算法滤除噪声,将局部阈值与Otsu算法相结合来实现对Data Matrix码图像的二值化处理;然后,利用腐蚀、膨胀等算法进一步滤除背景,采用Canny算法检测Data Matrix码边缘;最后,使用Hough变换提取L形区域,并进行旋转、插值等操作实现对二维码区域的识别.实验结果表明,本方法在复杂光照、复杂拍摄角度的条件下,可实现对Data Matrix码的完整提取与识别,对Data Matrix码在电能计量装置上的推广有重要意义. 展开更多
关键词 Datamatrix 数字图像处理 条码识别 局部阈值 HOUGH变换
原文传递
MatrixVB在测量数据处理中的应用 被引量:1
19
作者 林永秋 《福建建设科技》 2010年第3期68-71,共4页
本文介绍了MatrixVB的主要功能以及引用方式,并且重点介绍了在测量数据处理中比较重要的两个方面,即矩阵运算以及图形处理。通过本文的实例可以发现,在VB中加载了MatrixVB插件之后,使得程序代码编写的负担大大减轻。MATLAB和VB具有较好... 本文介绍了MatrixVB的主要功能以及引用方式,并且重点介绍了在测量数据处理中比较重要的两个方面,即矩阵运算以及图形处理。通过本文的实例可以发现,在VB中加载了MatrixVB插件之后,使得程序代码编写的负担大大减轻。MATLAB和VB具有较好的互补性,采用MatrixVB和VB混合编程,既保留了MATLAB强大的矩阵运算图形处理等功能,又集成了VB高级语言界面友好易于编译等优点。说明了采用MatrixVB和VB混合编程在测量数据处理中具有较好的实用价值。 展开更多
关键词 matrixVB MATLAB VB 测量数据处理 矩阵运算 图形处理
在线阅读 下载PDF
HAZOP在化工工艺风险定量评价中的优化研究
20
作者 蔡奇琦 《当代化工研究》 2026年第1期179-181,共3页
化工生产过程当中包含着多种不同的风险源,传统HAZOP方法在进行风险量化评估时,存在主观性强且精度不高等问题。层次分析法能够为风险因素赋权提供相应的数学依据,模糊综合评价可以对风险等级进行量化处理,贝叶斯网络能够动态推演多节... 化工生产过程当中包含着多种不同的风险源,传统HAZOP方法在进行风险量化评估时,存在主观性强且精度不高等问题。层次分析法能够为风险因素赋权提供相应的数学依据,模糊综合评价可以对风险等级进行量化处理,贝叶斯网络能够动态推演多节点的风险概率情况。通过构建HAZOP-定量评价集成框架可将定性分析转化为可计算的风险数值矩阵,进而形成涵盖偏离识别、权重量化及概率推理的完整评价链条。石化装置应用情况表明该方法让风险等级判定一致性系数从0.64提升至0.89,高危节点漏判率从28%降低至11%,评价周期缩短了40%,为工艺安全管理提供了有效的量化支持,推动了风险评价从经验判断向数据驱动方向转变。 展开更多
关键词 HAZOP分析 风险定量评价 风险矩阵 概率推理 化工工艺安全
在线阅读 下载PDF
上一页 1 2 138 下一页 到第
使用帮助 返回顶部