衬套模型对商用车驾驶室载荷分解精度的影响研究Study on Influence of Bushing Model on Load Decomposition Accuracy of Commercial Vehicle Cab
芦伟;李伟;张艳玲;
摘要(Abstract):
针对商用车驾驶室疲劳载荷分解过程中衬套建模精度不足的问题,研究了基于神经网络和样条插值的衬套建模方法,并与衬套刚度试验结果进行对比,结果表明,基于神经网络的衬套模型在随机波形试验数据测试集上的精度提升较样条插值模型更明显。基于以上两种衬套模型分别建立驾驶室多体动力学模型,采用虚拟迭代法提取驾驶室疲劳载荷,在短波路工况上进行验证,发现基于神经网络的衬套建模方法的载荷分解精度较基于样条插值的建模方法提高了8.41%,且两种衬套建模方法都满足工程需要。
关键词(KeyWords): 神经网络;样条插值;衬套建模;载荷分解
基金项目(Foundation):
作者(Authors): 芦伟;李伟;张艳玲;
DOI: 10.19620/j.cnki.1000-3703.20190272
参考文献(References):
- [1]黄德惠,向建东,张吉平,等.货车驾驶室悬置系统性能参数优化设计[J].汽车技术,2018(12):33-37.
- [2]于增亮,张立军,余卓平.橡胶衬套力学特性半经验参数化模型[J].机械工程学报,2010,46(14):115-123.
- [3]于增亮,张立军,罗鹰.一种新的橡胶衬套半经验动力学模型[J].汽车技术,2010(8):6-11.
- [4]左曙光,朱俊兴,吴旭东,等.一种考虑粘弹塑性的新型橡胶衬套高频建模方法研究[J].制造业自动化,2014,36(16):24-29.
- [5]吴赵佳,侯永平,张建文.随机振动条件下的橡胶衬套疲劳寿命预测[J].汽车技术,2017(3):24-28.
- [6]ANDREW J B.Accurate Models for Complex Vehicle Components Using Empirical Methods[J]//SAE Technical Paper,2000-01-1625,2000.
- [7]ANDREW J B,THOMAS E R,SHAWN Y,et al.Predicting Tire Handling Performance Using Neural Network Models[J]//SAE Technical Paper,2004-01-1574,2004.
- [8]刘竞一,董强强,杨建森.某商用车驾驶室疲劳载荷提取及验证[J].机械设计与制造,2017(8):187-193.
- [9]王陶,王良模,LI T,等.基于真实路谱再现的商用车驾驶室疲劳破坏[J].华中科技大学学报(自然科学版),2017,45(5):61-66.
- [10]SJ?BERG M,KARI L.Non-Linear Behavior of a Rubber Isolator System Using Fractional Derivatives[J].Vehicle System Dynamics,2002,37(3):217-236.
- [11]SJ?BERG M,KARI L.Nonlinear Isolator Dynamics at Finite Deformations:An Effective Hyperelastic,Fractional Derivative,Generalized Friction Model[J].Nonlinear Dynamics,2003,33(3):323-336.
- [12]SJ?BERG M.Rubber Isolators-Measurements and Modelling Using Fractional Derivatives and Friction[J]//SAE Technical Paper,2000-01-3518,2000.
- [13]GARCíA-TáRRAGO M J,KARI L.Frequency and Amplitude Dependence of the Axial and Radial Stiffness of Carbon-black Filled Rubber Bushings[J].Polymer Testing,2007(26):629-638.
- [14]周志华.机器学习[M].北京:清华大学出版社,2016:97-98.
- [15]KOHONEN T.An Introduction to Neural Computing[J].Neural Networks,1988,1(1):3-16.
- [16]Scikit-learn.Multi-layer Perceptron Classifier[DB/OL].(2018-08-03).http://scikit-learn.org/dev/modules/generated/sklearn.neural_network.MLPClassifier.html.