Application of Valued Characteristic Multi-granularity Model in Fault Diagnosis of Planetary Gearboxes
-
摘要: 传感器失灵、通讯迟滞或数据离散化等多种不确定因素会导致行星齿轮箱故障诊断信息不完备情况的发生,而现有的故障诊断方法已难于适用。为此,提出一种基于数据驱动量化特征多粒度模型的行星齿轮箱故障诊断方法。首先,采用数据驱动量化特征关系对行星齿轮箱的不完备故障诊断信息进行分析;其次,利用基于悲观数据驱动量化特征多粒度模型的属性约简算法提取故障诊断决策规则;最后,使用朴素贝叶斯分类器(Naive Bayesian classifier,NBC)推断待诊行星齿轮箱状态。实验研究表明,该方法可准确地判断实例间的不可分辨关系,降低计算复杂度,提高故障诊断准确率。Abstract: There are many uncertain factors that may result in incomplete diagnostic information of planetary gearboxes, such as sensor failures, communication lags and data discretization, etc. However, the existing methods are not suitable for fault diagnosis. Therefore, a fault diagnosis method of planetary gearboxes based on data-driven valued characteristic multi-granularity model is proposed. Incomplete fault diagnosis information of planetary gearbox is analyzed using data-driven valued characteristic relation. Then, the attribute reduction algorithm based on pessimistic data-driven valued characteristic multi-granularity model is employed to extract fault diagnosis decision rules. Finally, the Naive Bayesian classifier is used to determine planetary gearbox condition. The experimental results demonstrate that this method can accurately determine indiscernibility relation among cases, reduce computational complexity, and enhance fault diagnosis accuracy.
-
表 1 不完备故障诊断信息系统
U c1 c2 c3 d u1 2 ? 1 Y u2 1 1 2 Y u3 ? 2 1 Y u4 2 1 2 Y u5 2 ? 2 Y u6 3 1 1 Y u7 * 1 * N u8 1 2 1 N 表 2 行星齿轮箱故障诊断信息系统
U Kv Cv Pv KI CI W D u1 1 1 1 * 1 1 N u2 1 * 1 1 * 1 ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ u60 2 2 3 * 2 1 u61 1 1 3 * 2 2 F1 u62 2 * 2 1 1 2 ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ u120 2 2 2 2 2 2 u121 2 3 2 2 3 2 F2 u122 * 2 * * 2 2 ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ u180 3 3 3 3 * 2 u181 2 2 * 3 3 ? F3 u182 3 * 2 4 * ? ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ u240 3 2 1 3 3 ? u241 2 2 1 1 1 3 F4 u242 1 1 1 * 2 3 ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ u300 1 * 3 2 1 3 表 3 行星齿轮箱故障诊断决策规则
序号 决策规则 数量 支持实例 1 (Kv, 1)∧(Pv, 2)∧(W, 1)→N 2 u1, u11 2 (Pv, 3)∧(W, 1)→N 3 u2, u17, u33 3 (Kv, 1)∧(Cv, 1)→F1 2 u61, u71, 4 (KI, 1)∧(CI, 1)→F1 3 u62, u81, u97 5 (Kv, 2)∧(KI, 2)→F2 2 u121, u122 6 (Kv, 3)∧(Pv, 3)∧(Cv, 3)→F2 3 u128, u129, u161 7 (KI, 3)∧(CI, 3)→F3 1 u181 8 (Kv, 3)∧(KI, 4)→F3 3 u182, u199, u230 9 (Kv, 2)∧(Pv, 1)∧(W, 3)→F4 2 u241, u252 ⋮ ⋮ ⋮ ⋮ 115 (KI, 2)∧(W, 3)→F4 3 u237, u299, u300 表 4 故障诊断平均准确率
模型 规则数 实例数 正确数 准确率 乐观多粒度模型 134 300 283 94.33% 悲观多粒度模型 115 300 293 97.67% -
[1] Li Y B, Yang Y T, Li G Y, et al. A fault diagnosis scheme for planetary gearboxes using modified multi-scale symbolic dynamic entropy and mRMR feature selection[J]. Mechanical Systems and Signal Processing, 2017, 91:295-312 doi: 10.1016/j.ymssp.2016.12.040 [2] Lei Y G, Lin J, Zuo M J, et al. Condition monitoring and fault diagnosis of planetary gearboxes:a review[J]. Measurement, 2014, 48:292-305 doi: 10.1016/j.measurement.2013.11.012 [3] 李冠瑾, 刘文艺, 高钦武.一种交叉遗传优化MHW的风电机组微弱特征提取方法[J].机械科学与技术, 2017, 36(10):1594-1597 http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract6846.shtmlLi G J, Liu W Y, Gao Q W. A wind turbine's weak feature extraction method using cross genetic optimized MHW[J]. Mechanical Science and Technology for Aerospace Engineering, 2017, 36(10):1594-1597(in Chinese) http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract6846.shtml [4] Cheng Z, Hu N Q, Zhang X F. Crack level estimation approach for planetary gearbox based on simulation signal and GRA[J]. Journal of Sound and Vibration, 2012, 331(26):5853-5863 doi: 10.1016/j.jsv.2012.07.035 [5] Vanhollebeke F, Peeters P, Helsen J, et al. Large scale validation of a flexible multibody wind turbine gearbox model[J]. Journal of Computational and Nonlinear Dynamics, 2015, 10(4):041006 doi: 10.1115/1.4028600 [6] Chen J L, Zhang C L, Zhang X Y, et al. Planetary gearbox condition monitoring of ship-based satellite communication antennas using ensemble multiwavelet analysis method[J]. Mechanical Systems and Signal Processing, 2015, 54-55:277-292 doi: 10.1016/j.ymssp.2014.07.026 [7] Noll M C, Godfrey J W, Schelenz R, et al. Analysis of time-domain signals of piezoelectric strain sensors on slow spinning planetary gearboxes[J]. Mechanical Systems and Signal Processing, 2016, 72-73:727-744 doi: 10.1016/j.ymssp.2015.10.028 [8] Feng Z P, Liang M. Complex signal analysis for planetary gearbox fault diagnosis via shift invariant dictionary learning[J]. Measurement, 2016, 90:382-395 doi: 10.1016/j.measurement.2016.04.078 [9] Khazaee M, Ahmadi H, Omid M, et al. Classifier fusion of vibration and acoustic signals for fault diagnosis and classification of planetary gears based on Dempster-Shafer evidence theory[J]. Proceedings of the Institution of Mechanical Engineers, Part E:Journal of Process Mechanical Engineering, 2014, 228(1):21-32 doi: 10.1177/0954408912469902 [10] Grzymala-Busse J W, Clark P G, Kuehnhausen M. Generalized probabilistic approximations of incomplete data[J]. International Journal of Approximate Reasoning, 2014, 55(1):180-196 doi: 10.1016/j.ijar.2013.04.007 [11] Wang G Y, Guan L H, Wu W Z, et al. Data-driven valued tolerance relation based on the extended rough set[J]. Fundamenta Informaticae, 2014, 132(3):349-363 http://cn.bing.com/academic/profile?id=407b8e93d10b5d65fbcd98f41afaa4cc&encoded=0&v=paper_preview&mkt=zh-cn [12] Qian Y H, Liang J Y, Yao Y Y, et al. MGRS:a multi-granulation rough set[J]. Information Sciences, 2010, 180(6):949-970 doi: 10.1016/j.ins.2009.11.023 [13] Yu J, Bai M Y, Wang G N, et al. Fault diagnosis of planetary gearbox with incomplete information using assignment reduction and flexible naive Bayesian classifier[J]. Journal of Mechanical Science and Technology, 2018, 32(1):37-47 doi: 10.1007/s12206-017-1205-y