Fault detection method of machine tool bearing based on empirical modal analysis
-
摘要: 针对主流机床的电机主轴轴承外圈故障检测问题,提出1种利用机床主轴电机定子电流信号进行非接触式故障诊断的方法,利用经验模态分解(EMD)对机床电机非平稳定子电流信号进行分析。采用经验模态分解方法提取定子电流信号的本征模函数(IMF)应用于维格纳分布(WVD),得到故障信号的维格纳分布轮廓图,最终利用人工神经网络进行故障样本的模式识别,可有效检测机床主轴轴承外圈缺陷。试验结果表明,在不同负载条件下,基于经验模态分解的维格纳分布定子电流监测对外圈缺陷的故障检测和诊断具有准确率高、计算量小以及检测成本低等优点,具有一定的工程实用及推广价值。Abstract: In order to solve the problem of fault detection of the outer ring for the motor spindle bearing using in the mainstream machine, a non-contact fault diagnosis method using the stator current signal of the machine tool spindle motor was proposed. The empirical mode decomposition (EMD) was used to analyze the non-stationary stator current signal of the machine tool motor, and the eigenmode function (IMF) of the stator current signal was extracted by the empirical mode decomposition method and applied to the Wigner -Ville distribution (WVD) to obtain the fault signal. Finally, the artificial neural network was used for pattern recognition of fault samples, which can effectively detect defects in the outer ring of machine tool spindle bearings. The test results show that the stator current monitoring with Wigner distribution based on empirical mode decomposition has the advantages of high accuracy, small amount of calculation and low detection cost. It has certain engineering practical and popularization value.
-
Key words:
- machine tool bearing fault detection /
- EMD /
- WVD /
- artificial neural network
-
表 1 NSK 6205 Z轴承参数
类别 深沟球轴承 轴承名称 NSK 6205Z 滚珠数量N/个 9 内圈直径d/mm 25 外圈直径D/mm 52 滚珠直径Bd/mm 7.96 轴承的节圆直径Pd/mm 38.5 接触角α 0° 表 2 外圈轴承缺陷的机械特性频率
Hz 空载 半负载 满负载 $ {f}_{r} $ 49.653 48.81 47.641 $ {f}_{o} $ 178.756 175.73 171.51 表 3 在不同载荷条件下的外圈故障识别率和分析时间
载荷条件 识别率/(%) 分析时间/s 满负载 98.6 0.015 半载 99.2 0.025 空载 98.9 0.01 -
[1] 剡昌锋, 沈光飞, 师燚, 等. 滚动轴承外圈局部缺陷的有限元动力学分析[J]. 兰州理工大学学报, 2019, 45(2): 32-39. doi: 10.3969/j.issn.1673-5196.2019.02.006 [2] 李学东, 张云, 马晓莉. 基于振动分析的滚动轴承故障诊断系统设计[J]. 仪表技术与传感器, 2012(8): 71-73. doi: 10.3969/j.issn.1002-1841.2012.08.025 [3] 徐明林. 基于小波降噪和经验模态分解的滚动轴承故障诊断[D]. 哈尔滨: 哈尔滨工业大学. 2013. [4] Osman S, Wang W. A morphological Hilbert-Huang transform technique for bearing fault detection[J]. IEEE Transactions on Instrumentation & Measurement, 2016, 65(11): 2646-2656. [5] 李宏, 王崇武, 贺昱曜. 基于参数估计模型的对转永磁无刷直流电机实时故障诊断方法[J]. 西北工业大学学报, 2011, 29(5): 732-737. doi: 10.3969/j.issn.1000-2758.2011.05.014 [6] Yazici B, Kliman G B, Premerlani W J, et al. An adaptive, online, statistical method for bearing fault detection using stator current[C]. IEEE Industry Applications Society Annual Meeting. New Orleans, LA: IEEE, 1997: 213-220. [7] Jung J H, Lee J J, Kwon B H. Online diagnosis of induction motors using MCSA[J]. IEEE Transactions on Industrial Electronics, 2006, 53(6): 1842-1852. doi: 10.1109/TIE.2006.885131 [8] 杨江天, 赵明元, 张志强, 等. 基于定子电流小波包分析的牵引电机轴承故障诊断[J]. 铁道学报, 2013, 35(2): 32-36. doi: 10.3969/j.issn.1001-8360.2013.02.006 [9] 李洪, 郝豪豪, 孙云莲. 具有独立分量的经验模态分解算法研究[J]. 哈尔滨工业大学学报, 2009, 41(7): 245-248. doi: 10.3321/j.issn:0367-6234.2009.07.059 [10] 刘宗礼, 曹洁, 郝元宏. 一种新的特征提取方法及其在模式识别中的应用[J]. 计算机应用, 2009, 29(4): 1032-1035. [11] 柴凯, 张梅军, 黄杰, 等. 基于奇异值分解(SVD)差分谱降噪和本征模函数(IMF)能量谱的改进Hilbert-Huang方法[J]. 科学技术与工程, 2015, 15(9): 90-96. doi: 10.3969/j.issn.1671-1815.2015.09.015 [12] 包志强, 王美, 黄琼丹, 等. 经验模态分解神经网络的研究与应用[J]. 计算机工程与设计, 2021, 42(12): 3510-3515. doi: 10.16208/j.issn1000-7024.2021.12.028 [13] 单鑫, 董文洪. 多层前馈人工神经网络在装备故障诊断中的应用[J]. 兵工自动化, 2006(5): 40-41. doi: 10.3969/j.issn.1006-1576.2006.05.018