深度学习框架下数控机床运动误差溯因方法

详细信息    查看全文 | 推荐本文 |

英文篇名：Motion error tracing of NC machine tools based on deep learning framework 作者：余永维 ; 杜柳青 英文作者：Yu Yongwei;Du Liuqing;College of Mechanical Engineering, Chongqing University of Technology; 关键词：数控机床 ; 运动误差 ; 深度学习 ; 溯因方法 ; 神经网络 英文关键词：NC machine tools;;motion error;;deep learning;;tracing method;;neural network 中文刊名：YQXB 英文刊名：Chinese Journal of Scientific Instrument 机构：重庆理工大学机械工程学院; 出版日期：2019-01-15 出版单位：仪器仪表学报 年：2019 期：v.40 基金：国家自然科学基金项目面上项目(51775074);; 重庆市重点产业共性关键技术创新重点研发项目(cstc2017zdcy-zdyfX0066,cstc2017zdcy-zdyfX0073);; 重庆市基础研究与前沿探索项目(cstc2018jcyjAX0352)资助 语种：中文; 页：YQXB201901004 页数：7 CN：01 ISSN：11-2179/TH 分类号：31-37

摘要

为了快速准确地识别数控机床误差因素,提出了一种基于深度学习框架的运动误差智能溯因方法。首先,建立数控机床圆运动误差轨迹基准圆模型,生成理论样本解决深度学习大数据训练问题;然后,基于深度学习目标检测框架,提出数控机床运动误差溯因模型,构建深度卷积网络层自动提取圆运动轨迹特征,改进候选区域生成机制实现轨迹判别及区域生成;最后,模型的区域识别层准确识别误差轨迹类型,通过圆运动轨迹与误差源的映射关系实现对运动误差快速准确智能溯因。实验表明,本文方法可行且适应性好,识别溯源准确,正确率超过96%。
        To identify the error factors of NC machine tools quickly and accurately, an intelligent method of motion error tracing based on deep learning framework is proposed. First, the reference circle model of circular motion error trajectory is established, and a large number of theoretical samples are generated with the model. Then, the motion error tracing model based on Faster R-CNN framework is presented. The deep convolution network of the model is constructed to automatically extract the feature of circular track. The generation mechanism of the candidate region is improved to realize trajectory discrimination and region generation. Finally, the error trajectory type is identified accurately by the region recognition layer of the model. Motion error is traced intelligently and quickly according to mapping the circular motion trajectory to the error source. Experimental results show that this method is workable and adaptable, and the accuracy of traceability is higher than 96%.

引文

[1] 马军旭,赵万华,张根保.国产数控机床精度保持性分析及研究现状[J].中国机械工程,2015,26(22):3108- 3115.MA J X,ZHAO W H,ZHANG G B. Research status and analyses on accuracy retentivity of domestic CNC machine tools[J]. China Mechanical Engineering, 2015,26(22):3108- 3115.
    [2] NORIYUKI K,MASAOMI T, YU T, et al. Sensitivity analysis in ball bar measurement of three-dimensional circular movement equivalent to cone-frustum cutting in five-axis machining centers [J]. Journal of Advanced Mechanical Design, System, and Manufacturing,2013,7(3):317- 332.
    [3] SOICHI I,CHIAKI O,HISASHI O. Construction of an error map of rotary axes on a five-axis machining center by static R-test[J]. International Journal of Machine Tools & Manufacture,2011,51(3):190- 200.
    [4] AGUADO S,SAMPER D,SANTOLARIA J,et al. Identification strategy of error parameter in volumetric error compensation of machine tool based on laser tracker measurements[J]. International Journal of Machine Tools and Manufacture,2012,53(1):160- 169.
    [5] 姜忠,丁杰雄,王伟,等.基于RTCP功能的五轴数控机床动态误差溯源方法[J].机械工程学报,2016,52(7):187- 195.JIANG ZH,DING J X,WANG W, et al. Tracing the source of the dynamic error for five-axis CNC machine tool based on RTCP [J].Journal of Mechanical Engineering, 2016,52(7):187- 195.
    [6] 田文杰,牛文铁,常文芬,等.数控机床几何精度溯源方法研究[J].机械工程学报,2014,50(7):128- 135.TIAN W J,NIU W T,CHANG W F, et al. An identification approach for key geometric error sources of machine tool based on sensitivity analysis[J].Journal of Mechanical Engineering, 2014,50(7):128- 135.
    [7] 程强,刘广博,刘志峰,等.基于敏感度分析的机床关键性几何误差源识别方法[J].机械工程学报,2012,48(7):171- 179.CHEN Q, LIU G B, LIU ZH F, et al. An identification Approach for key geometric error sources of machine tool based on sensitivity analysis[J].Journal of Mechanical Engineering,2012, 48(7):171- 179.
    [8] 杜柳青,殷国富,余永维.基于图形识别的数控机床运动误差溯因方法[J].仪器仪表学报,2014,35(7):1662- 1668.DU L Q, YIN G F, YU Y W. Abduction of CNC machine tool′s motion error based on graphic recognition[J].Chinese Journal of Scientific Instrument,2014,35(7):1662- 1668.
    [9] ZAPATA J,VILAR R,RUIZ R. Performance evaluation of an automatic inspection system of weld defects in radiographic images based on neuroclassifiers[J].Expert Systems with Applications,2011,38(7):8812- 8824.
    [10] 杨理践,曹辉.基于深度学习的管道焊缝法兰组件识别方法[J].仪器仪表学报,2018,39(2):193- 202.YANG L J,CAO H. Deep learning based weld and flange identification in pipeline [J]. Chinese Journal of Scientific Instrument,2018,39(2):193- 202.
    [11] 赵光权,刘小勇,姜泽东,等.基于深度学习的轴承健康因子无监督构建方法[J].仪器仪表学报.2018,39(6):82- 88.ZHAO G Q,LIU X Y,JIANG Z D, et al. Unsupervised health indicator of bearing based on deep learning [J]. Chinese Journal of Scientific Instrument. 2018,39(6):82- 88.
    [12] 何怡刚,汪涛,施天成,等.基于传感器标签与深度学习的变压器状态监测方法研究[J].电子测量与仪器学报,2018,32(9):72- 79.HE Y G,WANG T,SHI T CH, et al. Research on monitoring technology of transformer based on RIFD sensor tag and deep learning[J]. Journal of Electronic Measurement and Instrument. 2018,32(9):72- 79.
    [13] 荣凡稳,郑伟,陈冉,等.基于深度学习的运动心率测量系统[J].电子测量与仪器学报,2017, 31(12):1912- 1917.RONG F W,ZHENG W, CHEN R, et al. Sportive heart rate measuring system based on deep learning[J]. Journal of Electronic Measurement and Instrumentation, 2017, 31(12): 1912- 1917.
    [14] 余永维,殷国富,殷鹰,等.基于深度学习网络的射线图像缺陷智能识别方法[J].仪器仪表学报,2014,35(9):92- 99.YU Y W,YIN G F, YIN Y,et al. Defect recognition for radiographic image based on deep learning network[J]. Chinese Journal of Scientific Instrument,2014,35(9):92- 99.
    [15] 曲建岭,余路,袁涛,等. 基于一维卷积神经网络的滚动轴承自适应故障诊断算法[J].仪器仪表学报.2018,39(7):134- 143.QU J L,YU L,YUAN T, et al. Adaptive fault diagnosis algorithm for rolling bearings based on one-dimensional convolutional neural network [J]. Chinese Journal of Scientific Instrument.2018,39(7):134- 143.
    [16] CHEN J, CHEN X L, YANG J, et al. Optimization of a training set for more robust face detection [J]. Pattern Recognition, 2009, 42 (11):2828- 2840.
    [17] SUN Y,WANG X G,TANG X O. Deep learning face representation from predicting 10,000 classes[C]. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2014:1891- 1898.
    [18] REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2015, 39 (6): 1137- 1149.