基于神经网络的薄壁件加工变形预测方法

doi:10. 3969/ j. issn. 1000-1093. 2013. 07. 007

兵工学报 ›› 2013, Vol. 34 ›› Issue (7): 840-845.doi: 10. 3969/ j. issn. 1000-1093. 2013. 07. 007

基于神经网络的薄壁件加工变形预测方法

秦国华, 张运建, 叶海潮

南昌航空大学无损检测教育部重点实验室, 江西南昌330063

收稿日期:2012-03-26 修回日期:2012-03-26 上线日期:2014-08-19
通讯作者: 秦国华 E-mail:qghwzx@126. com
作者简介:秦国华(1970—),男,教授,博士后。
基金资助:
国家自然科学基金项目(51165039);航空科学基金项目(2010ZE56014);江西省自然科学基金项目(2009GZC0104);江西省科技支撑计划重点项目(2010BGB00300);江西省省教育厅科学技术研究基金项目(GJJ10521);无损检测技术教育部重点实验室开放基金项目(ZD201029004)

A Neural Network-Based Prediction Method of Machining Deformation for Thin-walled Workpiece

QIN Guo-hua, ZHANG Yun-jian, YE Hai-chao

Key Laboratory of Nondestructive Testing of Ministry of Education, Nanchang Hangkong University, Nanchang 330063, Jiangxi, China

Received:2012-03-26 Revised:2012-03-26 Online:2014-08-19
Contact: QIN Guo-hua E-mail:qghwzx@126. com

摘要/Abstract

摘要： 在薄壁件的切削加工过程中,刀具几何结构是产生工件变形的重要因素之一。刀具单一角度所引起的工件变形规律可以很方便地通过有限元方法获得,但是,若同时考虑刀具多个角度的影响,仅仅利用有限元方法很难揭示多种工况与工件变形之间的关系。为此,针对薄壁件的铣削加工建立了三维有限元分析模型,通过实验数据与仿真值的比较验证有限元分析模型的有效性,以便利用有限元分析模型获取神经网络的训练样本;借助神经网络的非线性映射能力,通过有限的训练样本构建工件变形预测模型;将预测值与相应的有限元仿真值进行比较,结果表明预测误差在3%以内,进一步验证了建立的工件变形预测模型具有合理性。

关键词: 机械制造工艺与设备, 工件变形, 刀具结构, 有限元分析, BP 神经网络, 预测

Abstract: During the cutting operation of thin-walled workpiece, the tool parameter is an important factor causing the workpiece deformation. The deformation law of workpiece which is caused by the single tool angle can be obtained by finite element method. However, if multiple tool angles are synchronously considered only by using finite element method, the deformation law of workpiece is difficult to reveal. Therefore, the 3D finite element analysis model is established for the milling process of thin-walled work-piece. The comparison of the simulated values with the experimental results is carried out to validate the proposed finite element model. Thus, the viable finite element method can be used to obtain the training samples of neural network. And then, with the nonlinear mapping of neural network, the prediction model of workpiece deformation is suggested according to the finite training samples. Finally, the relative error in less than 3% of the predicted deformation to the corresponding simulated result shows that the proposed prediction model can be used to correctly obtain the workpiece deformation.

Key words: mechanical manufacturing process and equipment, workpiece deformation, tool structure, finite element analysis, BP neural network, prediction

中图分类号:

TH16

秦国华, 张运建, 叶海潮. 基于神经网络的薄壁件加工变形预测方法[J]. 兵工学报, 2013, 34(7): 840-845.

QIN Guo-hua, ZHANG Yun-jian, YE Hai-chao. A Neural Network-Based Prediction Method of Machining Deformation for Thin-walled Workpiece[J]. Acta Armamentarii, 2013, 34(7): 840-845.

参考文献

[1] 张臣, 周来水, 安鲁陵, 等. 刀具变形引起的球头铣刀加工误差建模[J]. 南京航空航天大学学报, 2008, 40(1): 94-99.

ZHANG Chen, ZHOU Lai-shui, AN Lu-ling, et al. Modeling of ball-end milling cutter deflection-induced machining error [ J].Journal of Nanjing University of Aeronautics and Astronautics,2008, 40(1): 94-99. (in Chinese)

[2] 梅中义, 高红, 王运巧. 飞机铝合金结构件数控加工变形分析与控制[J]. 北京航空航天大学学报, 2009, 35(2): 146-150.

MEI Zhong-yi, GAO Hong, WANG Yun-qiao. Analyzing and controlling distortion of aircraft aluminium alloy structural part in NC machining [J]. Journal of Beijing University of Aeronautics and Astronautics, 2009, 35(2): 146-150. (in Chinese)

[3] 唐东红, 孙厚芳, 王洪艳. 用BP 神经网络预测数控铣削变形[J]. 制造技术与机床, 2007, (8): 48-50.

TANG Dong-hong, SUN Hou-fang, WANG Hong-yan. Milling deformation forecast with BP neural network [ J]. Manufacturing Technology & Machine Tool, 2007, (8): 48-50. (in Chinese)

[4] 李目. 基于变形控制的薄壁件铣削加工参数优化及仿真研究[D]. 南京: 南京航空航天大学, 2008.

LI Mu. Research on the optimization of milling parameters and simulation of thin-walled parts based on the machining errors control [D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2008. (in Chinese)

[5] 刘新玲, 戚厚军. 基于神经网络的铣削复杂薄壁件受力变形分析和建模研究[J]. 机械制造, 2009, 47(535): 3-5.

LIU Xin-ling, QI Hou-jun. Forced deformation analysis and modeling research of milling of complex thin-walled parts based on neural network [J], Machinery, 2009, 47 (535): 3- 5. (in Chinese)

[6] Johnson R, Cook W K. A constitutive model and data for metals subjected to large strains high strain rates and high temperatures[C] //Proceedings of the 7th International Symposium on Balistics. Hague: IBC, 1983: 541-547.

[7] 张士林,任颂赞. 简明铝合金手册[M]. 上海: 上海科学技术出版社, 2001.

ZHANG Shi-lin, REN Song-zan. Brief handbook of aluminium alloy[ M]. Shanghai: Shanghai Science and Technology Press,2001. (in Chinese)

[8] 陈献廷. 硬质合金使用手册[M]. 北京: 冶金工业出版社,1986.

CHEN Xian-ting. Handbook of hard alloys [M]. Beijing: Metallurgical Industry Press, 1986. (in Chinese)

[9] 薛嘉庆. 最优化原理及方法[M]. 北京: 冶金工业出版社,1992.

XUE Jia-qing. Optimization principles and methods [M]. Beijing: Metallurgical Industry Press, 1992. (in Chinese)

[1]	励明君，姜毅，马立琦，潘霄. 基于神经网络算法的发射场坪承载能力预测方法[J]. 兵工学报, 2022, 43(5): 982-991.
[2]	陈军，张岳，陈晓威，佟龑. 基于模糊灰色认知图的复杂战场智能态势感知建模方法[J]. 兵工学报, 2022, 43(5): 1093-1106.
[3]	韩朝帅，诸雪征，顾进，张宏远，张赫，左钦文. 基于OpenFOAM的化学危害扩散预测求解器的开发与验证[J]. 兵工学报, 2022, 43(5): 1155-1166.
[4]	辛大钧，薛琨. 基于人工神经网络的非球形破片阻力系数预测模型[J]. 兵工学报, 2022, 43(5): 1083-1092.
[5]	许晓东，唐圣金，谢建，于传强，王凤飞，韩洋洋. 随机退化应力作用下设备剩余寿命预测方法[J]. 兵工学报, 2022, 43(3): 712-719.
[6]	张磊, 李世民，康淑瑰，王铁宁，郭猛超. 装甲装备维修器材需求数据模拟[J]. 兵工学报, 2022, 43(3): 720-728.
[7]	张军，温昊，刘志林，李国胜. 高速多体船的解析预测控制减纵摇方法[J]. 兵工学报, 2021, 42(9): 1962-1971.
[8]	陈路明，廖自力，马晓军，刘春光. 基于分层控制的混合动力车辆实时能量管理策略[J]. 兵工学报, 2021, 42(8): 1580-1591.
[9]	张明宇, 王琦, 于洋. 基于长短时记忆算法的热应力下半导体器件故障预测模型[J]. 兵工学报, 2021, 42(6): 1265-1274.
[10]	张杰，马晓军，刘春光，袁东，张运银. 双侧独立电驱动履带车辆反馈线性化解耦与预测行驶控制[J]. 兵工学报, 2021, 42(4): 697-705.
[11]	张永梅，赖裕平，马健喆，冯超，束颉. 基于视频的装甲车和飞机检测跟踪及轨迹预测算法[J]. 兵工学报, 2021, 42(3): 545-554.
[12]	马润波，董丽虹，王海斗. 基于支持向量机的热喷涂层接触疲劳寿命预测[J]. 兵工学报, 2021, 42(12): 2743-2752.
[13]	刘聪，刘辉，韩立金，陈科. 分布式电驱动车辆极限越野环境下高速避障与稳定性控制[J]. 兵工学报, 2021, 42(10): 2102-2113.
[14]	黄庆，倪大成，马长军，罗德荣，黄守道，李中启，程汪扬，姚超，邓海军，董平，陆雄建. 特种车辆电驱动系统的预测直接功率控制策略[J]. 兵工学报, 2021, 42(10): 2123-2129.
[15]	刘春光，陈路明，张运银，张征，徐浩轩. 基于改进灰色马尔可夫链的电传动装甲车辆负载需求功率预测[J]. 兵工学报, 2021, 42(10): 2130-2144.

基于神经网络的薄壁件加工变形预测方法

A Neural Network-Based Prediction Method of Machining Deformation for Thin-walled Workpiece

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