Fault Diagnosis of Check Valve for Diaphragm Pump with Multi-time Domain Feature and SVM
-
摘要: 针对隔膜泵单向阀故障机理分析不足的问题,提出了一种基于多时域特征与SVM的单向阀故障诊断方法。信号的时域特征是最早应用且最为简洁实用的特征参数,对单向阀的时域振动信号进行简要分析,并介绍几种信号时域指标与特征,根据单向阀振动信号的特点选取出三种时域指标与特征作为故障诊断的特征值;将特征值构成训练集输入到SVM分类器训练诊断模型;用测试样本进行故障诊断实验。实验证明,本文中提出的方法对高压隔膜泵单向阀的故障诊断准确率为98%,具有所需样本信号长度较短的优点。Abstract: Aiming at the insufficient analysis of the fault mechanism of the check (one-way) valve for the diaphragm pump, a new fault diagnosis method based on multi-time domain feature and SVM was proposed in this paper. The time domain feature of the signal is the earliest application and the most concise and practical feature parameter. Therefore, this paper briefly analyzes the time domain vibration signal of the one-way valve at first, and introduces several signal time domain indicators and characteristics, according to the characteristics of the one-way valve vibration signal. Three types of time domain indicators and features were selected as feature values for fault diagnosis. The training set of feature values was input into the SVM classifier training diagnosis model. The fault diagnosis experiment was performed with the test sample, and the experiment proved that the proposed method was applicable to high pressure diaphragm pumps. The fault diagnostic accuracy of the valve is 98%, and the method has the advantage of a shorter sample signal length.
-
Key words:
- check valve /
- fault detection /
- fault feature /
- SVM
-
表 1 信号采集器件型号
器件名称 型号 三缸曲轴活塞式隔膜泵 TZPM 振动加速度传感器 PCB-ICP 加速度校准器 PCB-394C06 高精度动态数据采集卡 PXIe-3342 控制器 PXI-3050EXT 2.7 Hz 工控机 PXI-9108EXT PXI机箱 表 2 部分样本的指标
样本类别 峭度值 近似熵值 裕度因子 0 11.993 4 0.784 2 27.055 9 0 9.703 0 0.726 6 26.595 4 0 5.877 3 0..943 8 20.611 4 0 6.824 7 0.886 5 19.510 0 0 4.537 5 0.953 9 14.634 0 0 12.097 5 0.870 1 29.605 4 0 6.327 9 0.808 4 16.017 0 0 7.096 1 0.953 1 20.078 8 0 8.299 3 0.760 7 22.247 4 0 6.286 2 0.954 6 27.657 5 1 2.607 5 1.197 9 3.054 9 1 4.995 0 0.980 4 7.247 6 1 2.756 8 1.200 6 3.848 2 1 6.395 1.020 2 7.260 0 1 4.039 1.195 3 4.149 6 1 6.229 0 0.957 6 7.247 2 1 3.550 6 1.072 9 5.044 9 1 4.180 5 1.121 6 7.338 2 1 3.436 4 1.174 8 3.794 5 1 2.547 4 1.190 7 2.390 8 表 3 两种核函数的识别准确率及误判率
核函数类别 准确率 误判率 Linear 92% 2% RBF 98% 0% 表 4 单一特征值进行识别的准确率及误判率
特征类别 准确率 误判率 峭度值 86% 8% 近似熵 76% 24% 裕度因子 92% 2% -
[1] 姜景升, 王华庆, 柯燕亮, 等.基于LTSA与K-最近邻分类器的故障诊断[J].振动与冲击, 2017, 36(11):134-139 http://d.old.wanfangdata.com.cn/Periodical/zdycj201711021Jiang J S, Wang H Q, Ke Y L, et al. Fault diagnosis based on LTSA and K-Nearest neighbor classifier[J]. Journal of Vibration and Shock, 2017, 36(11):134-139(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdycj201711021 [2] 戴云, 徐安, 陆孟和.基于粗糙集的语音情感分类的决策方法[J].统计与决策, 2008(15):56-58 http://d.old.wanfangdata.com.cn/Periodical/tjyjc200815018Dai Y, Xu A, Lu M H. Decision method of emotion based on rough set[J]. Statistics and Decision, 2008(15):56-58(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/tjyjc200815018 [3] 吴志清, 张文凡.Timoshenko梁自由振动分析及在称重仪中应用的研究[J].机床与液压, 2015, 43(11):169-172 doi: 10.3969/j.issn.1001-3881.2015.11.045Wu Z Q, Zhang W F. Analysis of free vibration of Timoshenko beam and study on its application in weighing instrument[J]. Machine Tool & Hydraulics, 2015, 43(11):169-172(in Chinese) doi: 10.3969/j.issn.1001-3881.2015.11.045 [4] 彭宵微, 彭翠云.基于无量纲参数分析的驱动机构运行性能研究[J].核电子学与探测技术, 2015, 35(10):1011-1013 doi: 10.3969/j.issn.0258-0934.2015.10.015Peng X W, Peng C Y. Research on the running performance of control rod drive mechanism based on dimensionless parameters analysis[J]. Nuclear Electronics & Detection Technology, 2015, 35(10):1011-1013(in Chinese) doi: 10.3969/j.issn.0258-0934.2015.10.015 [5] 张艳艳.近似熵理论及应用[J].中国医学物理学杂志, 2009, 26(6):1543-1546 http://d.old.wanfangdata.com.cn/Periodical/zgyxwlxzz200906020Zhang Y Y. The theory of approximate entropy and its application[J]. Chinese Journal of Medical Physics, 2009, 26(6):1543-1546(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zgyxwlxzz200906020 [6] 张淑清, 李威, 张立国, 等.基于多元经验模态分解互近似熵及GG聚类的轴承故障诊断[J].中国机械工程, 2016, 27(24):3362-3367 doi: 10.3969/j.issn.1004-132X.2016.24.017Zhang S Q, Li W, Zhang L G, et al. Bearing fault diagnosis based on multi-EMD, cApEn and GG clustering algorithm[J]. China Mechanical Engineering, 2016, 27(24):3362-3367(in Chinese) doi: 10.3969/j.issn.1004-132X.2016.24.017 [7] 刘畅, 伍星, 刘韬, 等.基于近似等距投影和支持向量机的滚动轴承故障诊断[J].振动与冲击, 2018, 37(5):234-239 http://d.old.wanfangdata.com.cn/Periodical/zdycj201805035Liu C, Wu X, Liu T, et al. Fault diagnosis of rolling bearings based on near-isometric projection and support vector machine[J]. Journal of Vibration and Shock, 2018, 37(5):234-239(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdycj201805035 [8] 殷俊, 杨万扣.核稀疏保持投影及生物特征识别应用[J].电子学报, 2013, 41(4):639-645 doi: 10.3969/j.issn.0372-2112.2013.04.003Yin J, Yang W K. Kernel sparsity preserving projections and its application to biometrics[J]. Acta Electronica Sinica, 2013, 41(4):639-645(in Chinese) doi: 10.3969/j.issn.0372-2112.2013.04.003 [9] 石志标, 陈斐, 曹丽华.基于EWT与排列熵的SVM汽轮机转子故障诊断[J].汽轮机技术, 2017, 59(6):439-442 doi: 10.3969/j.issn.1001-5884.2017.06.011Shi Z B, Chen F, Cao L H, Fault diagnosis of steam turbine rotor based on EWT and permutation entropy and SVM[J]. Turbine Technology, 2017, 59(6):439-442(in Chinese) doi: 10.3969/j.issn.1001-5884.2017.06.011 [10] 王强, 王莉, 沈进锐.基于多重分形与SVM的异步电动机轴承故障诊断[J].测控技术, 2017, 36(11):18-22 doi: 10.3969/j.issn.1000-8829.2017.11.004Wang Q, Wang L, Shen J R. Fault diagnosis of induction motor bearing based on multifractal and SVM[J]. Measurement & Control Technology, 2017, 36(11):18-22(in Chinese) doi: 10.3969/j.issn.1000-8829.2017.11.004 [11] 李艳军, 张建, 曹愈远, 等.基于模糊信息粒化和优化SVM的航空发动机性能趋势预测[J].航空动力学报, 2017, 32(12):3022-3030 http://d.old.wanfangdata.com.cn/Periodical/hkdlxb201712027Li Y J, Zhang J, Cao Y Y, et al. Forecasting of aero-engine performance trend based on fuzzy information granulation and optimized SVM[J]. Journal of Aerospace Power, 2017, 32(12):3022-3030(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/hkdlxb201712027