动能弹对混凝土靶侵彻深度的PSO-SVM预测

潘强; 张继春; 肖清华; 邹新宽; 石洪超

doi:10.11858/gywlxb.20170577

动能弹对混凝土靶侵彻深度的PSO-SVM预测

doi: 10.11858/gywlxb.20170577

潘强^1,,
张继春^1,*, ,,
肖清华¹,
邹新宽²,
石洪超^{1, 3}

1.
西南交通大学土木工程学院, 四川成都 610031
2.
自贡市城市建设投资开发集团有限公司, 四川自贡 643000
3.
成都工业学院建筑与环境工程系, 四川成都 610039

基金项目:

国家自然科学基金 50574076

详细信息

作者简介:
潘强(1985—), 男, 博士研究生, 主要从事爆炸与冲击防护研究.E-mail:pqiang1985@163.com

通讯作者:
张继春(1963—), 男, 博士, 教授, 主要从事工程爆破和爆炸防护研究.E-mail:jczhang2012@swjtu.edu.cn

中图分类号: O385;O24
计量
- 文章访问数: 7487
- HTML全文浏览量: 3090
- PDF下载量: 394
出版历程
- 收稿日期: 2017-05-05
- 修回日期: 2017-05-26

Prediction of Penetration Depth of Projectiles into Concrete Targets Based on PSO-SVM

1.
Institute of Civil Engineering, Southwest Jiaotong University, Chengdu 610031, China
2.
Zigong City Construction Investment Development Group Co.Ltd., Zigong 643000, China
3.
Department of Architectural and Environmental Engineering, Chengdu

摘要

摘要: 目前混凝土毁伤效应中侵彻深度的预测对防护工程设计与建设有着重要的指导意义, 传统的预测方法存在样本需求量大、预测误差大等问题。根据支持向量机原理, 采用粒子群算法优化模型参数, 提出了预测动能弹侵彻深度的粒子群-支持向量机方法, 并编写了相应的计算程序, 通过援引实测数据验证预测的准确性。结果表明:该方法对于小样本、非线性预测有较大优势, 相比于传统的灰色理论预测, 其预测相对误差较小(最大相对误差为3.18%); 随着训练样本量增多, 最大相对误差逐渐减小, 且变化速率逐渐减缓, 但计算量增大。因此, 粒子群-支持向量机方法用于动能弹侵彻混凝土靶体的深度预测是合理可行的。
- 粒子群优化 /
- 支持向量机 /
- 混凝土靶 /
- 侵彻深度 /
- 预测
Abstract: The prediction of the penetration depth of concrete in concrete damage effect is of great significance to the design and construction in protection engineering.However, the traditional methods for this prediciton involve such problems as requiring a great supply of samples, or suffering from a large prediction error, and so on.In this work, following the theory of the support vector machine (SVM) and according to the parameters optimized through the particle swarm optimization (PSO), the PSO-SVM for predicting the penetration depth was proposed.The corresponding programs were written and the prediction was verified by the experiment data.The results show that the PSO-SVM method has a great advantage for small samples and non-linear prediction.In comparison with the traditional grey theory, the relative predicted errors through the PSO-SVM method are smaller (the maximum relative error being 3.18%).As the number of the samples increases, the maximum relative errors decrease and the changing rate slows down whereas, however, the amount of calculation becomes larger.Above all, it is feasible to apply PSO-SVM method to the prediction of penetration depth of projectiles into concrete targets.
- particle swarm optimization /
- support vector machine /
- concrete targets /
- penetration depth /
- prediction

HTML全文

图 1 PSO-SVM模型构建流程图

Figure 1. Flow chart of PSO-SVM model building

下载: 全尺寸图片幻灯片

图 2 混凝土靶体侵彻的宏观破坏特征

Figure 2. Macroscopic damage features of concrete target by penetration

下载: 全尺寸图片幻灯片

图 3 实验、GM和PSO-SVM方法侵彻深度预测结果的比较

Figure 3. Contrast of penetration depths predicted by experiment, GM and PSO-SVM method

下载: 全尺寸图片幻灯片

表 1 侵彻深度预测结果对比

Table 1. Contrast table of predicted results of penetration depth

Samples	No.	σ_c/MPa	v/(m·s^-1)	s/m	s₁/m	δ₁/%	s₂/m	δ₂/%
Training samples	1	45	510	0.597	0.597	0	0.592	0.84
	2	23	510	0.834	0.742	11.03	0.839	-0.60
	3	45	612	0.716	0.816	-13.97	0.716	0
	4	23	612	1.001	1.089	-8.79	1.001	0
	5	45	680	0.795	0.791	0.50	0.799	-0.50
	6	23	680	1.113	1.129	-1.44	1.107	0.54
	7	45	748	0.875	0.856	2.17	0.880	-0.57
Testing samples	8	23	748	1.224	1.204	1.63	1.213	0.90
	9	45	850	0.994	0.971	2.31	1.001	-0.70
	10	23	850	1.391	1.320	5.10	1.368	1.65
	11	45	918	1.074	1.049	2.33	1.080	-0.56
	12	23	918	1.502	1.398	6.92	1.469	2.20
	13	45	1020	1.193	1.166	2.26	1.194	-0.08
	14	23	1020	1.669	1.514	9.29	1.616	3.18

下载: 导出CSV

表 2 不同训练样本数量下最大相对误差

Table 2. Maximum relative errors for different quantities of training samples

Training samples	Testing samples	Maximum relative error/%
4 (No.1-No.4)	10 (No.5-No.14)	10.02
6 (No.1-No.6)	8 (No.7-No.14)	5.43
7 (No.1-No.7)	7 (No.8-No.14)	3.18
8 (No.1-No.8)	6 (No.9-No.14)	2.85
10 (No.1-No.10)	4 (No.11-No.14)	1.18

下载: 导出CSV

参考文献(20)

[1]	沈俊, 刘瑞朝, 杨建超, 等.弹体侵彻岩体效应试验与理论研究[J].岩石力学与工程学报, 2008, 27(5):946-950. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yslxygcxb200805010 SHEN J, LIU R C, YANG J C, et al.Experimental and theoretical studies of projectiles penetrating rocks[J].Chinese Journal of Rock Mechanics and Engineering, 2008, 27(5):946-950. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yslxygcxb200805010
[2]	HEUZE F E.An overview of projectile penetration into geologic materials with emphasis on rock[J].International Journal of Rock Mechanics and Mining Sciences, 1991, 27(1):39-45. https://linkinghub.elsevier.com/retrieve/pii/014890629090003K
[3]	尹放林, 严少华, 钱七虎, 等.弹体侵彻深度计算公式对比研究[J].爆炸与冲击, 2000, 20(1):79-82. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=bzycj200001014 YIN F L, YAN S H, QIAN Q H, et al.Contrast analysis of calculation of penetration depth of projectiles[J].Explosion and Shock Waves, 2000, 20(1):79-82. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=bzycj200001014
[4]	FORRESTAL M J, FREW D J, HANCHAK S J, et al.Penetration of grout and concrete targets with ogive-nose steel projectiles[J].International Journal of Impacting Engineering, 1996, 18(5):465-476. doi: 10.1016/0734-743X(95)00048-F
[5]	YOUNG C W. Penetration equations: SAND 97-2426[R]. Albuquerque: Sandia National Laboratories, 1997.
[6]	BERNARD R S. Empirical analysis of projectile penetration in rock: AD-A047989[R]. 1977.
[7]	陈利, 张庆明.半穿甲战斗部的侵彻深度计算方法[J].北京理工大学学报, 2002, 22(4):162-165. http://industry.wanfangdata.com.cn/dl/Detail/Periodical?id=Periodical_bjlgdxxb200202008 CHEN L, ZHANG Q M.Methods of calculation of the penetration depth for semi-armor-piercing[J].Journal of Beijing Institute of Technology, 2002, 22(4):162-165. http://industry.wanfangdata.com.cn/dl/Detail/Periodical?id=Periodical_bjlgdxxb200202008
[8]	冷冰林, 许金余, 陈勇, 等.弹丸在不同速率下斜侵彻混凝土的数值模拟[J].弹箭与制导学报, 2008, 28(3):123-126. http://d.wanfangdata.com.cn/Periodical_djyzdxb200803038.aspx LENG B L, XU J Y, CHEN Y, et al.Numerical simulation of projection oblique penetration of concrete under different velocity[J].Journal of Projectiles, Rockets, Missiles and Guidance, 2008, 28(3):123-126. http://d.wanfangdata.com.cn/Periodical_djyzdxb200803038.aspx
[9]	陈星明, 刘彤, 肖正学.弹丸对混凝土靶体侵彻深度的灰色模拟与预测[J].高压物理学报, 2014, 28(2):221-226. doi: 10.11858/gywlxb.2014.02.014 CHEN X M, LIU T, XIAO Z X.Gray simulation and prediction on projectiles penetration depth of concrete targets [J].Chinese Journal of High Pressure Physics, 2014, 28(2):221-226. doi: 10.11858/gywlxb.2014.02.014
[10]	王玉岚, 李建光.弹体侵彻岩石深度的神经网络模型[J].电网与水力发电进展, 2007, 23(6):63-67. http://industry.wanfangdata.com.cn/dl/Detail/Periodical?id=Periodical_sxslfd200706016 WANG Y L, LI J G.Neural network model of penetration depth of projectiles into rock[J].Advances of Power System and Hydroelectric Engineering, 2007, 23(6):63-67. http://industry.wanfangdata.com.cn/dl/Detail/Periodical?id=Periodical_sxslfd200706016
[11]	刘艳, 许金余.灰色系统理论在混凝土侵彻深度计算中的应用[J].空军工程大学学报, 2006, 7(4):92-94. https://www.wenkuxiazai.com/doc/58893ce0f705cc1755270972.html LIU Y, XU J Y.Application of gray theory to determination of depth calculation of projectiles penetration into concrete[J].Journal of Air Force Engineering University, 2006, 7(4):92-94. https://www.wenkuxiazai.com/doc/58893ce0f705cc1755270972.html
[12]	熊岗, 陈章潮.灰色预测模型的缺陷及改进方法[J].系统工程, 1992, 10(6):32-36. http://www.wenkuxiazai.com/doc/92e7c720482fb4daa58d4bc5-2.html XIONG G, CHEN Z C.The shortcoming of grey forecasting model and a modified method for it[J].Systems Engineering, 1992, 10(6):32-36. http://www.wenkuxiazai.com/doc/92e7c720482fb4daa58d4bc5-2.html
[13]	张大海, 江世芳, 史开泉.灰色预测公式的理论缺陷及改进[J].系统工程理论与实践, 2002(8):140-142. http://www.docin.com/p-1234905346.html ZHANG D H, JIANG S F, SHI K Q.Theoretical defect of grey prediction formula and its improvement[J].Systems Engineering Theory and Practice, 2002(8):140-142. http://www.docin.com/p-1234905346.html
[14]	赵洪波.岩土力学与工程中的支持向量机分析[M].北京:煤炭工业出版社, 2008.
[15]	史峰, 王辉, 郁磊, 等.MATLAB智能算法30个案例分析[M].北京:北京航空航天大学出版社, 2011.
[16]	王小川, 史峰, 郁磊, 等.MATLAB神经网络43个案例分析[M].北京:北京航空航天大学出版社, 2013.
[17]	张树霞, 赵悍东, 韩志高.基于PSO-SVM的侵彻效果预测方法[J].中北大学学报, 2015, 36(2):166-170. http://www.cnki.com.cn/Article/CJFDTotal-HBGG201502016.htm ZHANG S X, ZHAO H D, HAN Z G.Method of penetrate result prediction based on PSO-SVM [J].Journal of North University of China, 2015, 36(2):166-170. http://www.cnki.com.cn/Article/CJFDTotal-HBGG201502016.htm
[18]	陈星明. 灰色系统理论在动能弹侵彻混凝土靶体中的应用研究[D]. 绵阳: 中国工程物理研究院, 2012. https://mall.cnki.net/lunwen-1012499433.html CHEN X M. Application study on penetration of projectile against concrete target about grey system theory[D]. Mianyang: Chinese Academy of Engineering Physics, 2012. https://mall.cnki.net/lunwen-1012499433.html
[19]	吝曼卿, 郭学彬, 肖正学, 等.弹性波CT在混凝土靶体冲击损伤实验中的应用[J].实验力学, 2010, 25(2):136-141. http://www.wenkuxiazai.com/doc/333afb8fcc22bcd126ff0cd5.html LIN M Q, GUO X B, XIAO Z X, et al.Application of elastic wave CT to impacting damage experiment of concrete targets[J].Journal of Experimental Mechanics, 2010, 25(2):136-141. http://www.wenkuxiazai.com/doc/333afb8fcc22bcd126ff0cd5.html
[20]	曹菊珍, 梁龙河, 王政. 高速碰撞中弹靶动态本构关系和力学特性的数值研究[R]. 绵阳: 中国工程物理研究院, 2010. CAO J Z, LIANG L H, WANG Z. Numerical study of dynamic constitutive and mechanical properties about high-speed collision[R]. Mianyang: Chinese Academy of Engineering Physics, 2010.