铜基复合材料分类判别方法研究

doi:10.3969/j.issn.1000-1093.2015.05.023

兵工学报 ›› 2015, Vol. 36 ›› Issue (5): 921-927.doi: 10.3969/j.issn.1000-1093.2015.05.023

铜基复合材料分类判别方法研究

马润波^1，2，杜建华²，许世蒙¹

（1装甲兵工程学院数学室，北京 100072； 2.装甲兵工程学院装备再制造技术国防科技重点实验室，北京 100072）

收稿日期:2014-07-15 修回日期:2014-07-15 上线日期:2015-07-09
通讯作者: 马润波 E-mail:13810470589@139.com
作者简介:马润波（1976—），女，讲师
基金资助:
国家自然科学基金项目（51001117）；北京市自然科学基金项目（3132024）

Research on Classification Discriminating Method of Copper Matrix Composites

MA Run-bo^1，2， DU Jian-hua²， XU Shi-meng¹

(1.Section of Mathematics, Academy of Armored Force Engineering, Beijing 100072， China；2.National Key Lab for Remanufacturing, Academy of Armored Forces Engineering, Beijing 100072， China)

Received:2014-07-15 Revised:2014-07-15 Online:2015-07-09
Contact: MA Run-bo E-mail:13810470589@139.com

摘要/Abstract

摘要： 对正交分割铜基复合材料表面微观形貌图，随机选取子图组成训练集。通过对基元模型的选取与设计，提取石墨颗粒的特征指标，并进行分布规律的统计推断，可知石墨颗粒的径心、斜率、分形维数和稠密度服从正态分布，长径、短径服从对数正态分布，经显著性检验可以推断石墨颗粒的分布规律具有一致性。进一步采用因子分析法，确定了表征铜基复合材料表面微观形貌复杂程度的主要指标，利用支持向量机原理，对不同石墨质量分数的铜基复合材料进行了分类判别。结果表明，通过表面微观形貌，对石墨质量分数为16%的铜基复合材料的分类准确率达到了83.333%，对石墨质量分数为10%与20%的两种铜基复合材料之间的分类准确率达到了100%.

关键词: 金属材料, 铜基复合材料, 基元, 分类, 支持向量机

Abstract: Surface microtopography of copper matrix composite is partitioned by orthogonal method, and the subgraphs are selected as training set at random. The characteristics of graphite particles are extracted by selecting and designing a basic element model, and the statistical inference of distribution law is made. It is known that the diameter center, slope, fractal dimension and density of graphite particles obey normal distribution, and its long axis and minor axis submit to logarithmic normal distribution. The distribution law of graphite particles can be inferred to be consistent by the way of significance test. The factor analysis method is used to determine the primary indices which characterize the complex surface microtopography of copper matrix composites, and the support vector machine principle is used to classify and discriminate the copper matrix composites with different graphite contents. The results show that, through the analysis of surface morphology, the classification accuracy of copper matrix composites, of which and mass fraction of graphite is 16%, reaches 83.333%, the classification accuracy of the two kinds of copper matrix composites, of which the mass fractions of graphite are 10% and 20%, between them reaches 100%.

Key words: metallic material, copper matrix composite, basic element, classification, support vector machine

中图分类号:

O212
TF125

马润波，杜建华，许世蒙. 铜基复合材料分类判别方法研究[J]. 兵工学报, 2015, 36(5): 921-927.

MA Run-bo， DU Jian-hua， XU Shi-meng. Research on Classification Discriminating Method of Copper Matrix Composites[J]. Acta Armamentarii, 2015, 36(5): 921-927.

参考文献

［1］马润波，许世蒙，杜建华，等.铜基复合材料组织形态分形特征的统计分析与研究[J]. 数学的实践与认识, 2012, 42(7):155-162.
MA Run-bo，XU Shi-meng，DU Jian-hua，et al. Statistical analysis and research on the microstructure fractal characteristics of copper matrix composites[J]. Mathematics in Practice and Theory, 2012, 42(7):155-162.（in Chinese）
[2] Han Y Y，Guo H，Yin F Z，et al. Microstructure and thermal conductivity of copper matrix composites reinforced with mixtures of diamond and sic particles[J]. Rare Metals, 2012,31(1):58-63.
[3] 史平安,莫军,郝志明,等.非均质材料界面损伤的数值模拟方法研究[J].兵器材料科学与工程,2008,31(4):9-14.
SHI Ping-an，MO Jun，HAO Zhi-ming，et al. Methods on numerical simulation of interfacial deboning in a heterogeneous material[J]. Ordnance Material Science and Engineering,2008,31(4):9-14.（in Chinese）
[4] 杜建华,冯建林,计德林，等.纳米SiO₂含量对铜基摩擦材料摩擦学性能的影响[J].粉末冶金技术,2008, 26(1):11-14.
DU Jian-hua，FENG Jian-lin，Ji De-lin，et al. Effects of the content of the nano-SiO₂ on the tribology performance of the Cu-based friction materials[J]. Powder Metallurgy Technology，2008, 26(1):11-14.（in Chinese）
[5] Corte S C，Vapnik V. Support vector networks [J]. Machine Learning, 1995, 20(3): 273-297.
[6] Vapnik V N. The nature of statistical learning theory [M].New York: Springer-Verlag, 1995.
[7] Suykens J A K，Vanewallel J. Least square support vector machine classifiers [J]. Neural Processing Letters，1999, 9(3):293-300.
[8] Chapelle O，Haffner P，Vapnik V N. Support vector machines for histogram-based image classification [J].IEEE Transactions on Neural Networks, 1999, 10(5): 1055-1064.
[9] Fan Z G，Wang K A，Lu B L. Feature selection for fast image classification with support vector machines[J]. Lecture Notes in Computer Science, 2004, 3316(1):1026-1031.
[10] Anindya G，Tarit G, Bhar R B. Classification of yarn interlacement pattern in fabrics using least square support vector machines [J]. Fibers and Polymers, 2013, 14(7): 1215-1219.
[11] Mandelbrot B B. The fractal geometry of nature [M].San Francisco: Freeman, 1982.
[12] Xu S M, Ma R B, Du J H,et al. The studies on basic element design of compound material surface structure models[J]. Applied Mechanics and Materials Equipment Environmental Engineering, 2013,365/366(II)：995-998.
[13] 冯秀,顾伯勤.表面形貌的研究现状及发展趋势[J]. 润滑与密封,2006,174(2): 168-170.
FENG Xiu，GU Bo-qin. The research status and development tendency of surface topography [J]. Lubrication Engineering,2006,174(2): 168-170.（in Chinese）
[14] 李水根.分形[M].北京：高等教育出版社，2004.
LI Shui-gen. Fractal[M]. Beijing：Higher Education Press,2004.（in Chinese）
[15] 王星.非参数统计[M].北京：清华大学出版社，2009.
WANG Xing. The parameter estimation[M].Beijing：Tsinghua University Press,2009.（in Chinese）
[16] 茆诗松, 王静龙, 濮晓龙. 高等数理统计[M].北京:高等教育出版社，1998.
MAO Shi-song，WANG Jing-long，PU Xiao-long. Higher mathematical statistics[M].Beijing：Higher Education Press,1998.（in Chinese）
[17] 黄玮，冯蕴雯，吕震宙.极小子样试验的虚拟增广样本评估方法[J].西北工业大学学报, 2005,23(3): 384-387.
HUANG Wei, FENG Yun-wen, LYU Zhen-zhou. Virtually expanded sample estimation method for extremely small-scale sample test[J]. Journal of Northwestern Polytechnical University, 2005,23(3): 384-387.（in Chinese）
[18] 邓乃扬, 田英杰. 数据挖掘中的新方法—支持向量机[M]. 北京：科学出版社，2004.
DENG Nai-yang,TIAN Ying-jie. New method in data mining: support vector machine[M].Beijing：Science Press,2004.（in Chinese）
[19] 安金龙，王正欧，马振平.一种新的支持向量机多类分类方法[J].信息与控制,2004,33(3):262-267.
AN Jin-long，WANG Zheng-ou，MA Zhen-ping. A new SVM multiclass classification method[J]. Information and Control,2004,33(3):262-267.（in Chinese）
[20] 张小云,刘允才.高斯核支撑向量机的性能分析[J].计算机工程，2003，29 (8)：22-25.
ZHANG Xiao-yun，LIU Yun-cai. Performance analysis of support vector machines with Gauss kernel[J]. Computer Engineering，2003，29(8)：22-25.（in Chinese）

[1]	王博洋, 关海杰, 龚建伟, 陈慧岩, 赵卉菁. 面向异构履带车辆的统一运动规划方法[J]. 兵工学报, 2022, 43(2): 241-251.
[2]	陆瑶敏, 龚建伟, 王博洋, 关海杰. 基于多重示范的智能车辆运动基元表征与序列生成[J]. 兵工学报, 2021, 42(4): 851-861.
[3]	李志农, 曾文钧, 闻庆松, 沈功田, 沈永娜. 静载拉伸和低周疲劳下Q235钢磁声发射特性[J]. 兵工学报, 2021, 42(1): 185-191.
[4]	王成，吴岩，杨廷飞. 利用改进单分类支持向量机提升舰船尾流目标的检测准确率[J]. 兵工学报, 2020, 41(9): 1887-1893.
[5]	王博洋，龚建伟，熊光明，张瑞增，陈慧岩，席军强. 有级转向履带车辆的驾驶员操控行为模型[J]. 兵工学报, 2020, 41(12): 2379-2388.
[6]	梁文利，陈慧岩，王博洋. 基于独立电驱动履带车辆的地面参量估计方法研究[J]. 兵工学报, 2019, 40(6): 1146-1153.
[7]	沙桂英，尹淼，韩恩厚，许道奎，徐永波，于爽. Mg-4.3Li-1.08Y-5.83Zn合金的微观组织与塑性不稳定性研究[J]. 兵工学报, 2017, 38(3): 534-539.
[8]	王博洋，龚建伟，高天云，陈慧岩，席军强. 基于高斯混合-隐马尔可夫模型的速差转向履带车辆横向控制驾驶员模型[J]. 兵工学报, 2017, 38(12): 2301-2308.
[9]	江伟华，童峰，王彬，刘世刚. 采用主分量分析的非合作水声通信信号调制识别[J]. 兵工学报, 2016, 37(9): 1670-1676.
[10]	张彪，闫晓鹏，栗苹，李志强，池庆玺. 基于支持向量机的无线电引信抗扫频式干扰研究[J]. 兵工学报, 2016, 37(4): 635-640.
[11]	胡勇，饶丽，邹天平. 等温热处理对AM60-1Si合金半固态组织的影响[J]. 兵工学报, 2016, 37(4): 690-695.
[12]	苗润，王伟力，宋杨. TC4钛合金动态II型裂纹扩展速度与加载速度的关系研究[J]. 兵工学报, 2016, 37(12): 2331-2339.
[13]	王绪虎，陈建峰，韩晶，张群飞. 单压差型矢量水听器方位估计的优化研究[J]. 兵工学报, 2014, 35(3): 340-346.
[14]	杨冬丽，王琳，杨杰，赵登辉. 泡沫铝复合结构的应力波防护性能研究[J]. 兵工学报, 2014, 35(1): 96-101.
[15]	郑宝明, 姜炜, 陈伟, 李凤生. 高活性 AlMg/ KMnO₄铝热剂的性能研究[J]. 兵工学报, 2013, 34(9): 1161-1166.

铜基复合材料分类判别方法研究

Research on Classification Discriminating Method of Copper Matrix Composites

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价