POLYGONAL WEAR IDENTIFICATION OF WHEELS BASED ON OPTIMIZED MULTIPLE KERNEL EXTREME LEARNING MACHINE
-
摘要: 多边形车轮是铁路机车车辆中普遍存在的一种磨损现象, 随着列车运营里程的增加, 车轮磨耗程度显著提升, 严重影响着列车乘坐舒适性和运营安全性, 借助于列车运营监测大数据开展多边形车轮动态检测方法研究具有重要意义. 本研究基于列车轴箱垂向加速度建立了多边形车轮定量识别模型, 首先通过阶次分析识别出轴箱加速度中包含的多边形车轮主要阶次, 同时获取各阶次对应的加速度幅值信息, 在此基础上引入加速度信号熵特征共同构建多边形车轮磨耗幅值识别特征矩阵, 然后建立遗传变异粒子群优化多核极限学习机 (GMPSO-MKELM) 识别模型, 通过特征矩阵与磨耗幅值的映射关系, 进一步实现了车轮多边形磨耗幅值识别. 通过仿真与现场实测数据研究结果表明, 所提出的识别模型能有效地从轴箱加速度中提取多边形车轮主要阶次, 磨耗幅值的识别精度均优于对比模型且具有较高的检测效率, 可实现均方根误差为0.0010 (仿真结果) 与0.0134 (试验结果) 的精确识别, 本文提出的多边形车轮磨耗识别模型可为列车车轮检测与智能维护提供理论基础.Abstract: Polygonal wheel is a common wear phenomenon in railway vehicles. With the increase in operating mileage, the degree of wheel wear increases significantly, which seriously affects the ride comfort and operation safety of the trains. It is of great significance to develop the polygonal wheel dynamic detection method with the help of monitoring big data. In this study, a polygonal wheel quantitative identification model is established based on the vertical acceleration signals of the axle box. Firstly, the main orders of the polygonal wheels contained in the acceleration signals of the axle box are identified through order analysis, and the acceleration amplitude corresponding to the main order is obtained at the same time. On this basis, the characteristic matrix of polygonal wheel wear amplitude identification is constructed by the acceleration amplitude of the main order and the entropy characteristic of the acceleration signal. Then the genetic mutation particle swarm optimized multiple kernel extreme learning machine (GMPSO-MKELM) identification model is established. Through the mapping between the characteristic matrix and the wear amplitude of the polygonal wheel, the wear amplitude identification is realized. The results of the simulation and field test show that the proposed model can effectively extract the main order of polygonal wheels from axle box acceleration. The identification accuracy of the proposed model is better than that of the comparison models in wear amplitude, and it has high detection efficiency. The proposed model can achieve accurate identification with a root mean square error of 0.0010 (simulation) and 0.0134 (field test). The proposed polygonal wheel identification model can provide a theoretical basis for the detection and intelligent maintenance of railway vehicle wheels.
-
Key words:
- polygonal wheel /
- axle box acceleration signal /
- GMPSO /
- MKELM /
- dynamic detection
-
图 8 多边形车轮磨耗幅值识别结果
Figure 8. The wear amplitude identification results of polygonal wheels
图 9 不同训练集样本数量的识别误差
Figure 9. Identification errors of the different number of training samples
图 10 现场试验: (a)试验机车; (b)轴箱加速度测量;(c)车轮镟修; (d)多边形车轮测量
Figure 10. Field testing: (a) testing locomotive; (b) measurement of axle box acceleration; (c) wheel lathing; (d) measurement of polygonal wheels
图 11 试验多边形车轮阶次与磨耗幅值测试结果
Figure 11. The order and wear amplitude testing results of polygonal wheels
图 14 多边形车轮磨耗幅值识别结果
Figure 14. The wear amplitude identification results of polygonal wheels
表 1 不同模型的磨耗幅值识别RMSE与时间
Table 1. Identification RMSE and time of wear amplitude with different models
Models Test 1 Test 2 Test 3 Time/s ELM 0.0259 0.0243 0.0266 0.0001 KELM (RBF kernel) 0.0107 0.0164 0.0091 0.0094 KELM (polynomial kernel) 0.0138 0.0149 0.0162 0.0029 KELM (wavelet kernel) 0.0037 0.0136 0.0048 0.0089 GMPSO-MKELM 0.0018 0.0037 0.0010 0.0173 
下载: 导出CSV
表 2 不同模型的磨耗幅值识别RMSE与时间
Table 2. Identification RMSE and time of wear amplitude with different models
Models Test 1 Test 2 Test 3 Time/s ELM 0.0242 0.0255 0.0268 0.0001 KELM (RBF kernel) 0.0225 0.0200 0.0235 0.0008 KELM (polynomial kernel) 0.0217 0.0231 0.0247 0.0003 KELM (wavelet kernel) 0.0210 0.0196 0.0202 0.0007 GMPSO-MKELM 0.0183 0.0134 0.0165 0.0018
下载: 导出CSV
-
[1] Tao G, Wen Z, Jin X, et al. Polygonisation of railway wheels: a critical review. Railway Engineering Science, 2020, 28(4): 317-345 doi: 10.1007/s40534-020-00222-x [2] Wang K, Yang Y, Yang Y, et al. An experimental investigation of the mechanism and mitigation measures for the coil spring fracture of a locomotive. Engineering Failure Analysis, 2022, in press [3] 王平, 汪鑫, 王源等. 基于高铁轨道不平顺的车轮不圆顺识别模型. 西南交通大学学报, 2020, 55(4): 681-687 + 678Wang Ping, Wang Xin, Wang Yuan, et al. Detection model of polygonal wheels based on track irregularity data in high-speed railway. Journal of Southwest Jiaotong University, 2020, 55(4): 681-687, 678 (in Chinese) [4] Chen S, Wang K, Peng Z, et al. Generalized dispersive mode decomposition: Algorithm and applications. Journal of Sound and Vibration, 2021, 492: 115800 doi: 10.1016/j.jsv.2020.115800 [5] Zheng F, Zhang B, Gao R, et al. A high-precision method for dynamically measuring train wheel diameter using three laser displacement transducers. Sensors, 2019, 19(19): 4148 doi: 10.3390/s19194148 [6] Zhang ZF, Gao Z, Liu YY, et al. Computer vision based method and system for online measurement of geometric parameters of train wheel sets. Sensors, 2012, 12(1): 334-346 [7] Liu F, Liang L, Xu G, et al. On-machine measurement and evaluation for the turned wheel profile of rail vehicle. IEEE Transactions on Instrumentation and Measurement, 2020, 70: 1-11 [8] Gao Y, Feng Q, Cui J. A simple method for dynamically measuring the diameters of train wheels using a one-dimensional laser displacement transducer. Optics and Lasers in Engineering, 2014, 53: 158-163 doi: 10.1016/j.optlaseng.2013.09.005 [9] Gong Z, Sun J, Zhang G. Dynamic measurement for the diameter of a train wheel based on structured-light vision. Sensors, 2016, 16(4): 564 doi: 10.3390/s16040564 [10] Sun Q, Chen C, Kemp AH, et al. An on-board detection framework for polygon wear of railway wheel based on vibration acceleration of axle-box. Mechanical Systems and Signal Processing, 2021, 153: 107540 doi: 10.1016/j.ymssp.2020.107540 [11] Chen S, Wang K, Chang C, et al. A two-level adaptive chirp mode decomposition method for the railway wheel flat detection under variable-speed conditions. Journal of Sound and Vibration, 2021, 498: 115963 doi: 10.1016/j.jsv.2021.115963 [12] 丁建明, 林建辉, 易彩等. 车轮不圆顺动态检测的时频特征圈内定位比较法. 振动与冲击, 2013, 32(19): 39-43 doi: 10.3969/j.issn.1000-3835.2013.19.008Ding Jianming, Lin Jianhui, Yi Cai, et al. Dynamic detection of out-of-round wheels using a comparison of time-frequency feature locatings. Journal of Vibration and Shock, 2013, 32(19): 39-43 (in Chinese) doi: 10.3969/j.issn.1000-3835.2013.19.008 [13] Song Y, Liang L, Du Y, et al. Railway polygonized wheel detection based on numerical time-frequency analysis of axle-box acceleration. Applied Sciences, 2020, 10(5): 1613 doi: 10.3390/app10051613 [14] 王其昂, 王腾, 倪一清. 基于卷积神经网络的高铁车轮损伤识别方法研究. 中国矿业大学学报, 2020, 49(4): 781-787Wang Qiang, Wang Teng, Ni Yiqing, et al. Damage detection of wheels for high-speed rail based on Convolutional Neural Network. Journal of China University of Mining & Technology, 2020, 49(4): 781-787 (in Chinese) [15] Chi Z, Zhou T, Huang S, et al. A data-driven approach for the health prognosis of high-speed train wheels. Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, 2020, 234(6): 735-747 doi: 10.1177/1748006X20929158 [16] Huang GB, Zhou H, Ding X, et al. Extreme learning machine for regression and multiclass classification. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, 2011, 42(2): 513-529 [17] 宋佳, 石若凌, 郭小红等. 基于核极限学习机的飞行器故障诊断方法. 清华大学学报(自然科学版), 2020, 60(10): 795-803Song Jia, Shi Ruoling, Guo Xiaohong, et al. KELM based diagnostics for air vehicle faults. Journal of Tsinghua University (Science and Technology), 2020, 60(10): 795-803 (in Chinese) [18] 李军, 李大超. 基于优化核极限学习机的风电功率时间序列预测. 物理学报, 2016, 65(13): 39-48Li Jun, Li Dachao. Wind power time series prediction using optimized kernel extreme learning machine method. Acta Physica Sinica, 2016, 65(13): 39-48 (in Chinese) [19] Hu Q, Qin A, Zhang Q, et al. Fault diagnosis based on weighted extreme learning machine with wavelet packet decomposition and KPCA. IEEE Sensors Journal, 2018, 18(20): 8472-8483 doi: 10.1109/JSEN.2018.2866708 [20] Cai Z, Gu J, Luo J, et al. Evolving an optimal kernel extreme learning machine by using an enhanced grey wolf optimization strategy. Expert Systems with Applications, 2019, 138: 112814 doi: 10.1016/j.eswa.2019.07.031 [21] Huang GB, Zhu QY, Siew CK. Extreme learning machine: theory and applications. Neurocomputing, 2006, 70(1-3): 489-501 doi: 10.1016/j.neucom.2005.12.126 [22] 张淑清, 苑世钰, 玉永等. 基于ICEEMD及AWOA优化ELM的机械故障诊断方法. 仪器仪表学报, 2019, 40(11): 172-180Zhang Shuqing, Yuan Shiyu, Yao Yuyong, et al. Machinery fault diagnosis method based on ICEMMD and AWOA optimized ELM. Chinese Journal of Scientific Instrument, 2019, 40(11): 172-180 (in Chinese) [23] Lu S, Gao W, Hong C, et al. A newly-designed fault diagnostic method for transformers via improved empirical wavelet transform and kernel extreme learning machine. Advanced Engineering Informatics, 2021, 49: 101320 doi: 10.1016/j.aei.2021.101320 [24] 方一鸣, 赵晓东, 张攀等. 基于改进灰狼算法和多核极限学习机的铁水硅含量预测建模. 控制理论与应用, 2020, 37(7): 1644-1654Fang Yiming, Zhao Xiaodong, Zhang Pan, et al. Prediction modeling of silicon content in liquid iron based on multiple kernel extreme learning machine and improved grey wolf optimizer. Control Theory & Applications, 2020, 37(7): 1644-1654 (in Chinese) [25] 杨锡运, 关文渊, 刘玉奇等. 基于粒子群优化的核极限学习机模型的风电功率区间预测方法. 中国电机工程学报, 2015, 35(S1): 146-153Yang Xiyun, Guan Wenyuan, Liu Yuqi, et al. Prediction intervals forecasts of wind power based on PSO-KELM. Proceedings of the CSEE, 2015, 35(S1): 146-153 (in Chinese) [26] Ding J, Huang L, Xiao D, et al. GMPSO-VMD algorithm and its application to rolling bearing fault feature extraction. Sensors, 2020, 20(7): 1946 doi: 10.3390/s20071946 [27] Xu F, Tse PW. A method combining refined composite multiscale fuzzy entropy with PSO-SVM for roller bearing fault diagnosis. Journal of Central South University, 2019, 26(9): 2404-2417 doi: 10.1007/s11771-019-4183-7 [28] 邓飞跃, 唐贵基. 基于时间−小波能量谱样本熵的滚动轴承智能诊断方法. 振动与冲击, 2017, 36(9): 28-34Deng Feiyue, Tang Guiji. An intelligent method for rolling element bearing fault diagnosis based on time-wavelet energy spectrum sample entropy. Journal of Vibration and Shock, 2017, 36(9): 28-34 (in Chinese) [29] 杨逸凡, 凌亮, 杨云帆等. 重载机车车轮擦伤下的轮轨动态响应. 工程力学, 2020, 37(12): 213-219 doi: 10.6052/j.issn.1000-4750.2020.01.0033Yang Yifan, Ling Liang, Yang Yunfan, et al. Wheel/rail dynamic responses due to the wheel flat of heavy-haul locomotives. Engineering Mechanics, 2020, 37(12): 213-219 (in Chinese) doi: 10.6052/j.issn.1000-4750.2020.01.0033 [30] 杨云帆, 刘志强, 高贤波等. 电力机车车轮非圆化磨耗特征及其对轮轨动态冲击作用影响分析. 机械工程学报, 2021, 57(4): 130-139 doi: 10.3901/JME.2021.04.130Yang Yunfan, Liu Zhiqiang, Gao Xianbo, et al. Analysis on essential characteristics of the polygonal wear of locomotive wheels and its effect on wheel/rail dynamic impact. Journal of Mechanical Engineering, 2021, 57(4): 130-139 (in Chinese) doi: 10.3901/JME.2021.04.130 [31] Zhai WM, Wang KY, Cai CB. Fundamentals of vehicle-track coupled dynamics. Vehicle System Dynamics, 2009, 47(11): 1349-1376 doi: 10.1080/00423110802621561 -
下载:
点击查看大图
图(14) / 表(2)
计量
- 文章访问数: 504
- HTML全文浏览量: 201
- PDF下载量: 94
- 被引次数: 0
