Indirect measurement method for wheel axle lateral force based on KF-LSTM

PDF(4829 KB)

Journal of Vibration and Shock ›› 2024, Vol. 43 ›› Issue (13) : 254-267.

SUN Zhaoyi, CHEN Jianzheng, WU Yue, XIE Qinglin

Author information +

History +

Abstract

The wheelset lateral force is an important indicator to evaluate the safety of the metro and it is significantly important to monitor the state online. An indirect measurement model of wheelset lateral force is developed by Kalman filter & Long short-term memory (KF-LSTM) algorithm. First, the process and observation equation of the Kalman filter are established by using the 17-degree-of-freedom lateral dynamics equation for the metro, and then the optimal observation variable is constructed for indirect measurement, and the long short-term memory network is used to measure the lateral force of the metro under the poor measurement results, the formula is corrected and compensated, and the validity of the KF-LSTM model is proved by numerical simulation and field trial. The result shows that: the wheelset lateral force can be accurately measured in the range of 0～20 Hz by KF-LSTM model. The correlation coefficient between the predicted value of the wheelset lateral force and the real value by simulation is about 0.85, and the average absolute error value is about 4.82 kN; The correlation coefficient between the predicted value and the force measured by instrument wheelset is about 0.84, and the average absolute error value is about 2.99 kN, and the warning standard is set while the metro is running on the line based on the method which provides practical guidance in engineering.

Key words

wheelset lateral force / Kalman filter / LSTM / indirect measurement

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

SUN Zhaoyi, CHEN Jianzheng, WU Yue, XIE Qinglin. Indirect measurement method for wheel axle lateral force based on KF-LSTM[J]. Journal of Vibration and Shock, 2024, 43(13): 254-267

References

[1] BRABIE D. On the influence of rail vehicle parameters on the derailment process and its consequences[D]. Stockholm: KTH Royal Institute of Technology, 2005. [2] 张志宏, 张宏, 陈有, 等. 基于遗传神经网络的履带行驶系统载荷识别方法[J]. 振动与冲击, 2022, 41(3): 54-61, 89. ZHANG Zhihong, ZHANG Hong, CHEN You, et al. Load identification method of track driving system based on genetic neural network[J]. Journal of Vibration and Shock, 2022, 41(3): 54-61, 89. [3] UHL T. The inverse identification problem and its technical application[J]. Archive of Applied Mechanics, 2007, 77: 325-337. [4] RONASI H, JOHANSSON H, LARSSON F. Identification of wheel–rail contact forces based on strain measurements, an inverse scheme and a finite‒element model of the wheel[J]. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 2014, 228(4): 343-354. [5] XIA F J, COLE C, WOLFS P. Grey box–based inverse wagon model to predict wheel‒rail contact forces from measured wagon body responses[J]. Vehicle System Dynamics, 2008, 46(S1): 469-479. [6] BERGGREN E G, LI M X D, SPANNAR J. A new approach to the analysis and presentation of vertical track geometry quality and rail roughness[J]. Wear, 2008, 265(9-10): 1488-1496. [7] 孙善超. 轨道—车辆系统轮轨力辨识及应用研究[D]. 北京: 中国铁道科学研究院, 2016. [8] 王明猛, 朱涛, 王小瑞, 等. 一种逆结构滤波法的轨道车辆轮轨力识别[J]. 振动工程学报, 2019, 32(4): 602-608. WANG Mingmeng, ZHU Tao, WANG Xiaorui, et al. An inverse structural filter method for wheel-rail contact force identification of railway vehicles[J]. Journal of Vibration Engineering, 2019, 32(4)：602-607. [9] LI Y F, LIU J X, WANG K Y, et al. Continuous measurement method of wheel/rail contact force based on neural network[C]//The 3rd International Conference on Transportation Engineering. Reston: America Society of Civil Engineers, 2011: 2533-2537. [10] URDA P, ACEITUNO J F, MUNOZ S, et al. Artificial neural networks applied to the measurement of lateral wheel–rail contact force: a comparison with a harmonic cancellation method[J]. Mechanism and Machine Theory, 2020, 153: 103968. [11] 黄小平, 王岩. 卡尔曼滤波原理及应用——MATLAB仿真[M]. 北京: 电子工业出版社, 2015. [12] 倪金福. 用卡尔曼滤波技术识别振动系统参数[J]. 振动与冲击, 1982(2): 23-36. NI Jinfu. The identification of vibration system parameters by the Kalman filtering technique[J]. Journal of Vibration and Shock, 1982(2): 23-36. [13] SARKKA S. Bayesian filtering and smoothing[M]. Cambridge: Cambridge University Press, 2013. [14] 翟婉明. 车辆—轨道耦合动力学[M]. 第四版. 北京: 科学出版社, 2015. [15] NEWMARK N M, ASCE F. A method of computation for structural dynamics [J]. Journal of the Engineering Mechanics Division, 1959, 85(3): 67-94. [16] 国家铁路局. 机车车辆动力学性能评定及试验鉴定规范: GB/T 5599-2019[S]. 北京: 中国标准出版社, 2019. [17] 马帅. 基于车辆响应的轨道几何状态评价方法研究[D].北京: 北京交通大学, 2020. [18] 陈晓丽, 陈光雄, 夏晨光, 等. 地铁轨道曲线半径与钢轨波磨的相关性研究[J]. 润滑与密封, 2021, 46(1): 124-129, 110. CHEN Xiaoli, CHEN Guangxiong, Xia Chenguang, et al. Study on the correlation between rail corrugation and curve radius of metro tracks[J]. Lubrication Engineering, 2021, 46(1): 124-129, 110. [19] 石怀涛, 尚亚俊, 白晓天, 等. 基于贝叶斯优化的SWDAE-LSTM滚动轴承早期故障预测方法研究[J]. 振动与冲击, 2021, 40(18): 286-297. SHI Huaitao, SHANG Yajun, BAI Xiaotian, et al. Early fault prediction method combining SWDAE and LSTM for rolling bearings based on Bayesian optimization[J]. Journal of Vibration and Shock, 2021, 40(18): 286-297. [20] 谢清林, 陶功权, 温泽峰. 基于一维卷积神经网络的地铁钢轨波磨识别方法[J]. 中南大学学报(自然科学版), 2021, 52(4): 1371-1379. XIE Qinglin, TAO Gongquan, WEN Zefeng. Detection method of metro rail corrugation based on 1-dimensional convolution neural network[J]. Journal of Central South University (Science and Technology), 2021, 52(4): 1371-1379. [21] 金学松, 张雪珊, 张剑, 等. 轮轨关系研究中的力学问题[J]. 机械强度, 2005, 27(4): 408-418. JIN Xuesong, ZHANG Xueshan, ZHANG Jian, et al. Mechanics in performance of wheel-rail[J]. Journal of Mechanical Strength, 2005, 27(4)：408-418. [22] 任愈, 陈建政. 测力轮对状态估计计算方法研究[J]. 振动与冲击, 2015, 34(9): 111-115. REN Yu, CHEN Jianzheng. Wheel/rail force calculation method based on state estimation[J]. Journal of Vibration and Shock, 2015, 34(9): 111-115. [23] 温泽峰, 金学松. 非稳态载荷下轮轨滚动接触及其钢轨波磨研究[J]. 摩擦学学报, 2007, 27(3)：252-257. WEN Zefeng, JIN Xuesong. Analysis of rolling contact and rail corrugation under non-steady stated loading[J]. Journal of Tribology, 2007, 27(3)：252-257. [24] 吴越, 韩健, 刘佳, 等. 高速列车车轮多边形磨耗对轮轨力和转向架振动行为的影响[J]. 机械工程学报, 2018, 54(4): 37-46. WU Yue, HAN Jian, LIU Jia, et al. Effect of high-speed train polygonal wheels on wheel/rail contact force and bogie vibration[J]. Journal of Mechanical Engineering, 2018, 54(4)：37-46.