吉林大学学报(工学版) ›› 2022, Vol. 52 ›› Issue (2): 329-337.doi: 10.13229/j.cnki.jdxbgxb20211079

• 车辆工程·机械工程 • 上一篇    

考虑工况变化的数控刀架运行状态异常检测方法

胡炜1,2,3(),陈传海1,2,郭劲言1,2(),刘志峰1,2,申桂香1,2,于春明4   

  1. 1.吉林大学 数控装备可靠性教育部重点实验室,长春 130022
    2.吉林大学 机械与航空航天工程学院,长春 1300225
    3.瑞典皇家理工学院,瑞典 斯德哥尔摩 25175
    4.沈阳机床股份有限公司,沈阳 110142
  • 收稿日期:2021-10-22 出版日期:2022-02-01 发布日期:2022-02-17
  • 通讯作者: 郭劲言 E-mail:weihu19@mails.jlu.edu.cn;gjy_jlu@163.com
  • 作者简介:胡炜(1994-),女,博士研究生.研究方向:数控装备可靠性评估与状态监测.E-mail:weihu19@mails.jlu.edu.cn
  • 基金资助:
    国家自然科学基金项目(51975249);吉林省重点科技攻关项目(20190302017GX)

An Anomaly detection method for numerical control turrets considering working conditions

Wei HU1,2,3(),Chuan-hai CHEN1,2,Jin-yan GUO1,2(),Zhi-feng LIU1,2,Gui-xiang SHEN1,2,Chun-ming YU4   

  1. 1.Key Laboratory of CNC Equipment Reliability,Ministry of Education,Jilin University,Changchun 130022,China
    2.College of Mechanical and Aerospace Engineering,Jilin University,Changchun 130022,China
    3.KTH Royal Institute of Technology,Stockholm 25175,Sweden
    4.Shenyang Machine Tool Co. ,Ltd. ,Shenyang 110142,China
  • Received:2021-10-22 Online:2022-02-01 Published:2022-02-17
  • Contact: Jin-yan GUO E-mail:weihu19@mails.jlu.edu.cn;gjy_jlu@163.com

摘要:

针对数控刀架故障数据少、难获取,且运行数据随工况变化致使故障诊断困难的问题,提出了一种基于非故障数据并考虑工况变化的运行状态异常检测方法。该方法通过多元高斯分布模型和考虑工况变化的偏差特征建模,确定刀架运行状态异常数据的评判依据。首先,通过统计分析找到刀架在不同工作过程中的关键工况与信号特征;其次,分别选择线性回归、信息增益判别、广义回归神经网络方法建立工况与信号特征的关系模型,得到观测信号特征与给定信号特征之间的偏差;最后,采用刀架正常状态时的数据训练模型。通过大量工况变化实验与异常模拟实验,并与传统多元高斯分布模型比较,得出本文所提模型能排除工况变化的影响,并能更好地识别异常状态。

关键词: 异常检测, 工况, 数控刀架, 多元高斯分布

Abstract:

The difficulties of failure data collection and operation data changeability hinder the application of fault diagnosis methods to turrets. Hence, an anomaly detection method using non-failure data and considering the change of working conditions was proposed for detecting turrets’ anomaly state during operation. The method studied the judgment principle of abnormal data through the multivariate Gaussian distribution(MGD) and the deviation characteristic associated with working conditions. First, the key working conditions and signal characteristics in different turret working processes were determined through statistical analysis. Second, some methods like linear regression, information gain, and generalized regression neural network were selected to model their relationships, respectively. Following that, the deviation of observation from the given signal characteristics is calculated. Finally, the operation data from turret normal state were used to train the model. Many experiments under different working conditions and abnormal simulation were conducted to verify that the proposed model can eliminate the influence of working conditions on abnormal judgment compared to the traditional MGD model.

Key words: anomaly detection, working condition, numerical control turret, multivariate Gaussian distribution

中图分类号: 

  • TG659

图1

传统MGD的工作原理"

图2

切削动作过程中信号特征与工况的关系模型"

图3

实验图"

表1

松开锁紧动作内工况与特征关系模型的拟合优度"

关系模型

F检验

p值)

t-检验(p值)决策 系数R
系数1系数2
油压-特征Ask0.000.000.000.764
油压-特征Asj0.000.000.020.858
油压-特征Bsk0.000.000.000.961
油压-特征Bsj0.000.000.000.848

图4

松开锁紧动作过程中信号特征的拟合结果"

图5

转位动作内信号特征的拟合结果"

表2

转位动作内工况与特征关系模型的拟合优度"

关系模型F检验 (p值)t-检验(p值)决策系数R
系数1系数2系数3
式(23)0.000.000.00-0.985
式(24)0.000.000.000.000.851

图6

切削动作内已知工况下拟合振动信号"

表3

各动作内所提模型参数结果"

动作信号特征μSε
松开AB[ 5.4×10-11 7.1×10-11[8.5×10-5 -1.3×10-4 ; -1.3×10-4 1.7×10-30.4460
锁紧AB[ 1.3×10-10 -2.0×10-17[3.7×10-3 -2.2×10-4 ; -2.2×10-4 3.4×10-30.0766
转位DE[ 1.4×10-10 -1.1×10-17[5.4×10-4 -9.3×10-3 ; -9.3×10-3 23.64]0.4725
切削GH[-7.4×10-5 0.4774][7.8×10-5 2.7×10-5 ; 2.7×10-5 1.3036]0.0003

图7

异常检测结果"

1 杨兆军, 陈传海, 陈菲, 等. 数控机床可靠性技术的研究进展[J].机械工程学报, 2013, 49(20): 130-139.
Yang Zhao-jun, Chen Chuan-hai, Chen Fei, et al. Progress in the research of reliability technology of machine tools[J]. Journal of Mechanical Engineering, 2013, 49(20): 130-139.
2 张英芝, 郑珊, 申桂香, 等. 采用重要度和模糊推理的数控刀架危害性分析[J]. 吉林大学学报: 工学版, 2012, 42(5): 1157-1162.
Zhang Ying-zhi, Zheng Shan, Shen Gui-xiang, et al. Criticality analysis for CNC turret adopting importance and fuzzy reasoning[J]. Journal of Jilin University(Engineering and Technology Edition), 2012, 42(5): 1157-1162.
3 陈斌, 陈松灿, 潘志松, 等. 异常检测综述[J]. 山东大学学报: 工学版, 2009, 39(6): 13-23.
Chen Bin, Chen Song-can, Pan Zhi-song, et al. Survey of outlier detection technologies[J]. Journal of Shandong University(Engineering Science), 2009, 39(6): 13-23.
4 周东华, 魏慕恒, 司小胜. 工业过程异常检测、寿命预测与维修决策的研究进展[J]. 自动化学报, 2013, 39(6): 711-722.
Zhou Dong-hua, Wei Mu-heng, Si Xiao-sheng. A survey on anomaly detection, life prediction and maintenance decision for industrial processes[J]. Acta Automatica Sinica, 2013, 39(6): 711-722.
5 丁建伟. 数据驱动的复杂装备异常检测方法[D]. 北京: 清华大学计算机科学与技术学院, 2015.
Ding Jian-wei. Data-driven complex equipment anomaly detection methods[D]. Beijing: College of Computer Science and Technology, Tsinghua University, 2015.
6 Ding Nan, Ma Hao-xuan, Gao Huan-bo, et al. Real-time anomaly detection based on long short-Term memory and Gaussian mixture model[J]. Computers & Electrical Engineering, 2019, 79: 106458.
7 Li G Y, Lee H, Rai A, et al. Analysis of operational and mechanical anomalies in scheduled commercial flights using a logarithmic multivariate Gaussian model[J]. Transportation Research, 2020, 110(1): 20-39.
8 Zeng X J, Yang M, Bo Y F. Gearbox oil temperature anomaly detection for wind turbine based on sparse Bayesian probability estimation[J]. International Journal of Electrical Power & Energy Systems, 2020, 123: 106233.
9 Zhao Hong-shan, Liu Hui-hai, Hu Wen-jing, et al. Anomaly detection and fault analysis of wind turbine components based on deep learning network[J]. Renewable Energy, 2018, 127(11): 825-834.
10 Sarmadi H, Karamodin A. A novel anomaly detection method based on adaptive mahalanobis-squared distance and one-class kNN rule for structural health monitoring under environmental effects[J]. Mechanical Systems and Signal Processing, 2020, 140(1): 106495.
11 谢群亚. 基于故障树的数控刀架可靠性分配方法研究[D]. 长春: 吉林大学机械与航空航天工程学院, 2017.
Xie Qun-ya. Research on reliability allocation method of CNC servo turret based fault tree analysis[D]. Changchun: School of Mechanical and Aerospace Engineering, Jilin University, 2017.
12 杨兆军, 杨川贵, 陈菲, 等. 基于最小二乘算法和SVDUKF算法的电液伺服加载优化[J]. 吉林大学学报: 工学版, 2014, 44(2): 392-397.
Yang Zhao-jun, Yang Chuan-gui, Chen Fei, et al. Optimization of the electro-hydraulic servo loading based on least square and SVDUKF algorithms[J]. Journal of Jilin University(Engineering and Technology Edition), 2014, 44(2): 392-397.
13 胡炜. 面向数控刀架可靠性试验的CPS研究[D]. 长春: 吉林大学机械与航空航天工程学院, 2019.
Hu Wei. Research on CPS of NC turret reliability test[D]. Changchun: School of Mechanical and Aerospace Engineering, Jilin University, 2019.
[1] 董绍江,朱朋,裴雪武,李洋,胡小林. 基于子领域自适应的变工况下滚动轴承故障诊断[J]. 吉林大学学报(工学版), 2022, 52(2): 288-295.
[2] 侯春萍,赵春月,王致芃,田海瑞. 基于有效异常样本构造的视频异常检测算法[J]. 吉林大学学报(工学版), 2021, 51(5): 1823-1829.
[3] 曾小华,宋美洁,宋大凤,王越. 基于车联网信息的公交客车行驶工况数据处理方法[J]. 吉林大学学报(工学版), 2021, 51(5): 1692-1699.
[4] 陈魏,雷雨龙,李兴忠,付尧,扈建龙,侯利国. 低速工况下渐开线圆柱直齿轮齿面粘着磨损计算[J]. 吉林大学学报(工学版), 2021, 51(5): 1628-1634.
[5] 李洪雪,李世武,孙文财,李玮,郭梦竹. 重型危险品半挂列车行驶工况的构建[J]. 吉林大学学报(工学版), 2021, 51(5): 1700-1707.
[6] 宋大凤,杨丽丽,曾小华,王星琦,梁伟智,杨南南. 基于行驶工况合成的混合动力汽车电池寿命优化[J]. 吉林大学学报(工学版), 2021, 51(3): 781-791.
[7] 李小雨,许男,仇韬,郭孔辉. 各向异性刚度对轮胎力学特性及车辆操纵性的影响[J]. 吉林大学学报(工学版), 2020, 50(2): 389-398.
[8] 刘玉梅,乔宁国,庄娇娇,刘鹏程,胡婷,陈立军. 基于多传感器数据融合的轨道车辆齿轮箱异常检测[J]. 吉林大学学报(工学版), 2019, 49(5): 1465-1470.
[9] 周华,杨志刚,朱晖. 基于整车风洞试验的MIRA车型数值计算[J]. 吉林大学学报(工学版), 2019, 49(4): 1043-1053.
[10] 林学东, 江涛, 许涛, 李德刚, 郭亮. 高压共轨柴油机起动工况高压泵控制策略[J]. 吉林大学学报(工学版), 2018, 48(5): 1436-1443.
[11] 宋大凤, 吴西涛, 曾小华, 杨南南, 李文远. 基于理论油耗模型的轻混重卡全生命周期成本分析[J]. 吉林大学学报(工学版), 2018, 48(5): 1313-1323.
[12] 曹骞, 李君, 刘宇, 曲大为. 基于马尔科夫链的长春市乘用车行驶工况构建[J]. 吉林大学学报(工学版), 2018, 48(5): 1366-1373.
[13] 张曼, 施树明. 典型汽车运行工况的状态转移特征分析[J]. 吉林大学学报(工学版), 2018, 48(4): 1008-1015.
[14] 于赫, 秦贵和, 孙铭会, 闫鑫, 王璇喆. 车载CAN总线网络安全问题及异常检测方法[J]. 吉林大学学报(工学版), 2016, 46(4): 1246-1253.
[15] 秦大同, 詹森, 漆正刚, 陈淑江. 基于K-均值聚类算法的行驶工况构建方法[J]. 吉林大学学报(工学版), 2016, 46(2): 383-389.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!