基于EKF-BP网络的矿用自卸车轮胎材料参数辨识

摘要
图/表
参考文献(19)
相关文章 (0)

全文: PDF (2073 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要轮胎材料参数对轮胎有限元模型至关重要，但轮胎材料多，结构复杂，导致轮胎材料参数难以获取，对此提出利用扩展卡尔曼（EKF）优化的BP神经网络辨识轮胎材料参数的方法。基于轮胎有限元模型，模拟了轮胎脉冲工况动态仿真，将仿真得到的轮胎垂向加速度作为网络理想输入样本，将需要辨识的轮胎材料参数作为网络理想输出样本，通过网络训练，建立两者之间的非线性映射网络模型。将经过小波去噪的轮胎垂向加速度试验数据输入训练好的网络，有效辨识出了轮胎材料参数。通过材料参数辨识的轮胎模型在相应工况下的仿真数据与试验数据的对比，显示两者最大误差为6.45%，证明了基于材料参数辨识的轮胎有限元模型垂向特性的准确性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张菲菲1 谷正气1
	2 张沙1 马骁骙1 朱一帆1

关键词 ：矿用自卸车轮胎；参数辨识；BP神经网络；扩展卡尔曼；小波去噪

Abstract：Tire material parameters on tire finite element model is very important, but the tire material is more and the structure is complex,leading to the tire material parameters are difficult to obtain, the method of using BP neural network which is optimized by Extend Kalman filter to identify the tire parameters is proposed. The dynamic simulation of the tire pulse condition is simulated based on the tire finite element model, The tire acceleration that is obtained by simulating is regarded as an ideal input sample of neural network,and the tire material parameters that is need to identify is regarded as an ideal output sample of neural network, then the nonlinear mapping network model between them is built by network training. The tire vertical acceleration test data which are denoised by wavelet input trained network, can effectively identify tire material parameters. Through the comparison of simulation data under corresponding conditions and the experimental data, show that the biggest error is 6.45%, it is proved that the accuracy of the vertical characteristics of tire finite element model based on the identification of material parameters.

Key words： Mining dump truck tire Parameter identification BP neural network Extend Kalman filter Wavelet denoise

收稿日期: 2015-06-16 出版日期: 2016-08-25

引用本文:

张菲菲1 谷正气1,2 张沙1 马骁骙1 朱一帆1. 基于EKF-BP网络的矿用自卸车轮胎材料参数辨识[J]. 振动与冲击, 2016, 35(17): 71-76.
Zhang Feifei 1 Gu Zhengqi 1,2 Zhang Sha1 Ma Xiaokui1 Zhu Yifan1 . Tire Material Parameters Identification of Mining Dump Truck Based on EKF-BP. JOURNAL OF VIBRATION AND SHOCK, 2016, 35(17): 71-76.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2016/V35/I17/71

[1] 石琴, 陈无畏, 洪洋, 等. 基于有限元理论的轮胎刚度特性的仿真研究[J]. 系统仿真学报, 2006, 18(6): 1445-1449.
     Shi Qin,Chen Wuwei,Hong Yang,etc.The Simulation of tire’s stiffness characteristics using finite element model[J]. Journal of System Simulation, 2006, 18(6): 1445-1449.
[2] 张德丰.MATLAB神经网络应用设计[M]. 北京:机械工业出版社,2012.
     ZhangDengfeng.MATLAB neural network application design[M].BeiJing:Mechanical industry press,2012.
[3] Yam J Y F, Chow T W S. A weight initialization method for improving training speed in feedforward neural network [J]. Neurocomputing, 2000, 30(1): 219-232.
[4] 谢延敏,王新宝. 基于灰色理论和GA-BP 的拉延筋参数反求[J]. 机械工程学报,2013,49(4):44-50.
     Xie Yanmin,Wang Xinbao.Parameter Inverse Problem for Drawbeads Based on the Gray Theory and GA-BP[J].Journal of Mechanical Engineering, 2013,49(4):44-50.
[5] Pichler B, Mang H A. Parameter Identification Based on First Order Approximation Neural Networks[J]. PAMM, 2003, 2(1): 440-441.
[6] Guarneri P, Rocca G, Gobbi M. A neural-network based model for the dynamic simulation of the tire/ suspension system while traversing road irregularities [J]. Neural Networks, IEEE Transactions on, 2008, 19(9): 1549-1563.
[7] 于德介,雷慧.一种基于神经网络的结构参数识别方法.湖南大学学报,1999,26(4)．
     Yu Dejie,Lei Hui.A Method for Structural Parameters Identification Using Neural Networks.Journal of Hunan University,1999，26(4)．
[8] Simon D. Training radial basis neural networks with the extended Kalman filter[J]. Neurocomputing, 2002, 48(1):455-475.
[9] 胡林, 谷正气, 黄晶, 等. 30.00 R51 子午线轮胎刚
    度仿真[J].系统仿真学报, 2008, 20(8): 2210-2214.
    Hu Lin,Gu Zhengqi,Huang Jing,etc. 30.00 R51 radial tire stiffness simulation [J]. Journal of System Simulation, 2008, 20(8): 2210-2214.
[10] 程冬冬. 子午线轮胎三维精细网格有限元分析[D].
      东华大学, 2014.
      Cheng Dongdong. Radial tire dimensional fine mesh finite element analysis [D]. Donghua University, 2014.
[11] 蒋霖, 文鸿雁. 基于 EKF 的神经网络在变形预测中的应用[J]. 桂林工学院学报, 2006, 26(1): 66-68.
     JIANG Lin,WEN Hong-yan.Application of neural network based on the extended Kalman filter to deformation prediction[J].Journal of Guilin University of Technology, 2006, 26(1): 66-68.
[12] 付梦印, 邓志红, 张继伟. Kalman 滤波理论及其在导航系统中的应用[M]. 科学出版社, 2003.
     Fu Mengyin,Deng Zhihong,Zhang Jiwei.Kalman filtering theory and its application in the navigation system[M]. Science press, 2003.
[13] 庄继德. 汽车轮胎学[M]. 北京理工大学出版社, 1996.
     Zhuang Jideng.Automobile tire Science[M].Beijing university of technology press, 1996.
[14] 薛雪. 车辆轮胎与土壤接触变形的有限元分析[D]. 西北农林科技大学, 2010.
     Xue Xue.Fea of vehicle tire-soil contact deformation [D]. Northwest A& F University, 2010.
[15] 程冬冬. 子午线轮胎三维精细网格有限元分析[D]. 东华大学, 2014.
     Cheng Dongdong.Three-dimensional fine mesh finite element analysis of radial tire[D]. DongHua University,2014.
[16] Hickernell F. A generalized discrepancy and quadrature error bound[J]. Mathematics of Computation of the American Mathematical Society, 1998, 67(221): 299-322.
[17] Brito N S D, Souza B A, Pires F A C. Daubechies wavelets in quality of electrical power[C]//Harmonics and Quality of Power Proceedings, 1998. Proceedings. 8th International Conference On. IEEE, 1998, 1: 511-515.
[18] 郭亚. 振动信号处理中的小波基选择研究 [D][J]. 合 BE: 合肥工业大学硕士论文, 2003.
     Guo Ya.Research on Wavelet Base Selection For Vibration Signal Proeessing[D][J].Hefei University of Technology, 2003.
[19] 姜永胜, 王其东. 汽车振动信号后处理的研究 [D][J]. 合肥工业大学学报, 2006.
     Jiang Yongsheng.Research on Post Processing for Vehicle Vibration Signal[D][J].Hefei University of Technology, 2006.