基于拉普拉斯特征马氏距离的滚珠丝杠健康评估

doi:10.3901/JME.2017.15.125

机械工程学报 ›› 2017, Vol. 53 ›› Issue (15): 125-130.doi: 10.3901/JME.2017.15.125

扫码分享

基于拉普拉斯特征马氏距离的滚珠丝杠健康评估

赵帅¹, 黄亦翔¹, 王浩任¹, 刘成良¹, 刘晓², 梁鑫光²

1. 上海交通大学机械系统与振动国家重点实验室上海 200240;
2. 上海航天设备制造总厂上海 200245

收稿日期:2016-06-11 修回日期:2016-12-26 出版日期:2017-08-05 发布日期:2017-08-05
通讯作者: 黄亦翔(通信作者),男,1980年出生,博士,讲师。主要研究方向为机电系统智能维护、信号处理与模式识别、特征提取与降维。E-mail:huang.yixiang@sjtu.edu.cn
作者简介:赵帅,男,1993年出生。主要研究方向为机电系统监控与智能维护、特征提取与降维。E-mail:yidingdou@sjtu.edu.cn
基金资助:
国家自然科学基金（51305258）、国家科技支撑计划（2014BAD08B01）国家科技重大专项（2014ZX04015-021）资助项目。

Laplacian Eigenmaps and Mahalanobis Distance based Health Assessment Methodology for Ball Screw

ZHAO Shuai¹, HUANG Yixiang¹, WANG Haoren¹, LIU Chengliang¹, LIU Xiao², LIANG Xinguang²

1. State Key Laboratory of Mechanical System and Vibration, Shanghai Jiao Tong University, Shanghai 200240;
2. Shanghai Aerospace Equipment Manufacturer, Shanghai 200245

Received:2016-06-11 Revised:2016-12-26 Online:2017-08-05 Published:2017-08-05

摘要/Abstract

摘要： 滚珠丝杠的性能是影响数控机床加工精度的重要因素之一。提出一种机床滚珠丝杠的性能衰退及健康状态的评估方法，该方法结合拉普拉斯特征降维与马氏距离分析模型，建立不同健康状态下传感器信号样本点在特征空间中与健康值的非线性映射关系，从而得到滚珠丝杠性能衰退程度的量化评估。通过不同健康状态的滚珠丝杠性能试验，将该方法应用于滚珠丝杠的驱动电动机速度和转矩信号，通过对传感器信号的内蕴流形及其不同健康状态下的采样特征点在内蕴特征空间中分布相关性分析，以得到量化的性能评估结果。与常见方法所得结果相比，该模型能准确地反映滚珠丝杠的性能衰退趋势，鲁棒性更好。该方法可采用数控机床自带的传感器，无须改动机床整体结构，不影响其动态加工性能，可广泛应用于工业数控机床滚珠丝杠的在线实时健康状态评估。

关键词: 滚珠丝杠, 健康评估, 拉普拉斯特征值映射, 马氏距离, 性能衰退

Abstract: The performance of ball screw is one of the important factors which affect the CNC machining precision. A method to assess the performance degradation and health status of ball screws is proposed based on the combination of Laplacian eigenmaps and Mahalanobis distance analysis to establish a nonlinear mapping relationship between the characteristics of the signals and health status of the ball screw. The health values are calculated to represent the degree of the performance degradation. Experiments have been conducted by testing the ball screws of different health status. The proposed method is performed on the speed and torque signals from the drive motor for validation. Compared with the results from the traditional dimensionality reduction methods, the results show that the proposed model is more accurate and robust. The model is featured with both the correlation analysis from the Mahalanobis distance analysis and the manifold learning analysis from the method of Laplacian eigenmaps. In addition, the method can be performed on the build-in sensors of the CNC machine tools so that there is no need to change the original structure design to avoid to the potential interferences with the dynamic processing performance, which enables a wide industrial applications of the online real-time health status assessment for the ball screws of the CNC machine tools.

Key words: ball screw, health evaluation, Laplacian eigenmaps, Mahalanobis distance, performance degradation

中图分类号:

TH17

赵帅, 黄亦翔, 王浩任, 刘成良, 刘晓, 梁鑫光. 基于拉普拉斯特征马氏距离的滚珠丝杠健康评估[J]. 机械工程学报, 2017, 53(15): 125-130.

ZHAO Shuai, HUANG Yixiang, WANG Haoren, LIU Chengliang, LIU Xiao, LIANG Xinguang. Laplacian Eigenmaps and Mahalanobis Distance based Health Assessment Methodology for Ball Screw[J]. Journal of Mechanical Engineering, 2017, 53(15): 125-130.

参考文献

[1] FENG Guohua, PAN Yilu. Investigation of ball screw preload variation based on dynamic modeling of a preload adjustable feed-drive system and spectrum analysis[J]. International Journal of Machine Tools & Manufacture,2012,52(1):85-96.
[2] KIM M S,CHUNG S C. Friction identification of ball-screw driven servomechanisms through the limit cycle analysis[J]. Mechatronics,2006,16(2):131-140.
[3] HUNG J P,LIN C Y,LUO T L. Fault detection of linear guide preload of a positioning stage with vibration-acoustic analysis[J]. Journal of Failure Analysis and Prevention,2011,11(6):684-692.
[4] 刘栋,梅雪松,冯斌,等. 基于Symlets小波滤波的滚珠丝杠伺服进给系统频响特性辨识[J]. 机械工程学报,2011,47(13):153-159. LIU Dong,MEI Xuesong,FENG Bin,et al. Frequency response identification for ballscrew servo driven system based on symlets wavelet[J]. Journal of Mechanical Engineering,2011,47(13):153-159.
[5] 赵敏,高宏力,许明恒,等. 多变量灰色模型在滚珠丝杠剩余寿命预测中的应用[J]. 计算机集成制造系统,2011,17(4):846-851. ZHAO Min,GAO Hongli,XU Mingheng,et al. Application of multi-variable grey model for ball screw remaining life prediction[J]. Computer Integrated Manufacturing Systems,2011,17(4):846-851.
[6] 张筱辰,高宏力,黄海凤,等. 数控机床滚珠丝杠副性能衰退评估技术[J]. 计算机集成制造系统,2015,21(5):1293-1300. ZHANG Xiaochen,GAO Hongli,HUANG Haifeng,et al. Performance degradation assessment technology for screw of NC machine tool[J]. Computer Integrated Manufacturing Systems,2015,21(5):1293-1300.
[7] 高宏力,李登万,许明恒. 基于人工智能的丝杠寿命预测技术[J]. 西南交通大学学报,2010,45(5):685-691. GAO Hongli,LI Dengwan,XU Mingheng. Intelligent monitoring system for screw life evaluation[J]. Journal of Southwest Jiaotong University,2010,45(5):685-691.
[8] 徐卫晓,谭继文,战红. 改进DST在数控机床滚珠丝杠副智能故障诊断中的应用[J]. 制造技术与机床,2014(9):163-166. XU Weixiao,TAN Jiwen,ZHAN Hong. Applications of improved D-S evidence theory in CNC machine ball screw intelligent fault diagnosis[J]. Manufacturing Technology & Machine Tool,2014(9):163-166.
[9] CHEN Chun,ZHANG Lijun,BU Jiajun,et al. Constrained Laplacian eigenmap for dimensionality reduction[J]. Neurocomputing,2010,73(4):951-958.
[10] LIN Jinshan,CHEN Qian. Fault diagnosis of rolling bearings based on multifractal detrended fluctuation analysis and mahalanobis distance criterion[J]. Mechanical Systems and Signal Processing,2013,38(2):515-533.

[1]	葛磊, 倪豪, 郝云晓, 赵斌, 苑永亮, 权龙. 新型液压回转-机械直线执行器的死区补偿控制[J]. 机械工程学报, 2024, 60(8): 337-347.
[2]	赵佳佳, 宋现春, 姜洪奎, 林明星, 宋丽伟. 考虑热弹流润滑作用的滚珠丝杠副摩擦力矩建模与分析[J]. 机械工程学报, 2023, 59(3): 176-188.
[3]	李月昊, 胡茑庆, 程哲, 尹正阳, 黄良远. 基于惯容的浮筏隔振系统人工自愈技术研究[J]. 机械工程学报, 2023, 59(18): 13-21.
[4]	郑锐, 周一帆. 针对比例风险退化系统健康评估的迭代算法[J]. 机械工程学报, 2023, 59(10): 366-373.
[5]	黄涛, 杜双江, 张钎, 曹华军. 滚珠丝杠传动系统线性变参数动力学模型试验建模方法[J]. 机械工程学报, 2022, 58(9): 98-106.
[6]	谢经伦, 周长光, 聂从辉, 冯虎田, 周华西. 双螺母滚珠丝杠副进给系统轴向固有频率衰退预测方法研究[J]. 机械工程学报, 2022, 58(11): 120-132.
[7]	黄文豪, 王军, 熊瑞, 陈欢, 王春. 温度和电压因素对超级电容性能衰退的影响机制及其老化建模研究[J]. 机械工程学报, 2022, 58(10): 235-244.
[8]	王建新, 李铁军, 朱军, 杜一锦. 低压开关柜运行在线监测和诊断系统设计[J]. 电气工程学报, 2021, 16(4): 189-195.
[9]	朱建新, 陈学东, 吕宝林, 王溢芳, 乔松, 陈嘉宏. 基于多维高斯贝叶斯的机械设备失效/故障智能诊断及参数影响分析[J]. 机械工程学报, 2020, 56(4): 35-41.
[10]	赵佳佳, 林明星, 宋现春, 骆伟超, 宋丽伟. 复合载荷下滚珠丝杠副全滚珠载荷分布建模与分析[J]. 机械工程学报, 2020, 56(17): 126-136.
[11]	李亚, 黄亦翔, 赵路杰, 刘成良. 基于t分布邻域嵌入与XGBoost的刀具多工况磨损评估[J]. 机械工程学报, 2020, 56(1): 132-140.
[12]	张康, 黄亦翔, 赵帅, 刘成良, 王吉云. 基于t-SNE数据驱动模型的盾构装备刀盘健康评估[J]. 机械工程学报, 2019, 55(7): 19-26.
[13]	朱林, 贾民平, 林明智, 陈晖. 融合应力集中因子与贝叶斯理论的结构健康评估方法[J]. 机械工程学报, 2019, 55(24): 21-27,36.
[14]	赵帅, 黄亦翔, 王浩任, 刘成良, 刘晓, 梁鑫光. 基于随机森林与主成分分析的刀具磨损评估[J]. 机械工程学报, 2017, 53(21): 181-189.
[15]	王恒, 张俊杰, 王永泉, 陈花玲. 考虑反向器影响的滚珠丝杠副摩擦力矩计算模型[J]. 机械工程学报, 2016, 52(7): 97-103.

基于拉普拉斯特征马氏距离的滚珠丝杠健康评估

Laplacian Eigenmaps and Mahalanobis Distance based Health Assessment Methodology for Ball Screw

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