基于混合神经遗传算法的扩胀管吸能特性预测
|
摘要
由于结构参数对吸能元件的吸能特性具有重要影响,将填充了泡沫铝的扩胀管与混合神经遗传算法相结合,对不同结构参数下扩胀管吸能特性进行分析预测。首先,基于泡沫铝填充机理设计出泡沫铝填充的扩胀式吸能装置,并建立有限元模型;然后利用非线性有限元软件LS-DYNA对不同参数变化情况下的扩胀管进行准静态轴向碰撞仿真;最后将胀管壁厚、诱导锥角、泡沫铝密度作为BP神经网络输入,扩胀管吸能特性参数作为网络输出,利用遗传算法优化网络权重和阈值,建立3层BP神经网络预测模型,经样本数据训练得到合适的网络。研究结果表明,网络预测值与期望值很接近,平均压溃载荷的误差值为3.02%,比吸能的误差值为4.82%,压缩力效率的误差值为0.92%,说明了该网络模型能够有效地预测扩胀管的吸能特性,并具有较高的精确度。
The structural parameters have an important influence on the energy absorption characteristics of energy absorbing devices,the expansion tube filled with aluminum foam is combined with the hybrid neural genetic algorithm,to analyze and predicate the energy absorption characteristics of expanded tube under different structural parameters. Firstly,Based on the mechanism of aluminum foam filling,the expansion energy absorption device of aluminum foam was designed,and the finite element model of the expanded tube was established;Then using the nonlinear finite element software LS-DYNA to simulate the quasi-static axial impact of the expansion tube under different parameters;Finally,the wall thickness of the expansion tube,the induced cone angle and the density of the foamed aluminum are used as the input of the BP neural network. The energy absorption characteristic parameters of expanding tube are used as network output,genetic algorithm is used to optimize network weights and thresholds. The 3 layer BP neural network prediction model is established,and the appropriate network is trained by the sample data. The results show that the predicted value is close to the expected value. The error value of the mean crushing load,specific energy absorption and compression efficiency are 3.02%,4.82% and 0.92%. It is shown that the network model can effectively predict the energy absorption characteristics of the expansion tube and has high accuracy.
The structural parameters have an important influence on the energy absorption characteristics of energy absorbing devices,the expansion tube filled with aluminum foam is combined with the hybrid neural genetic algorithm,to analyze and predicate the energy absorption characteristics of expanded tube under different structural parameters. Firstly,Based on the mechanism of aluminum foam filling,the expansion energy absorption device of aluminum foam was designed,and the finite element model of the expanded tube was established;Then using the nonlinear finite element software LS-DYNA to simulate the quasi-static axial impact of the expansion tube under different parameters;Finally,the wall thickness of the expansion tube,the induced cone angle and the density of the foamed aluminum are used as the input of the BP neural network. The energy absorption characteristic parameters of expanding tube are used as network output,genetic algorithm is used to optimize network weights and thresholds. The 3 layer BP neural network prediction model is established,and the appropriate network is trained by the sample data. The results show that the predicted value is close to the expected value. The error value of the mean crushing load,specific energy absorption and compression efficiency are 3.02%,4.82% and 0.92%. It is shown that the network model can effectively predict the energy absorption characteristics of the expansion tube and has high accuracy.
引文
[1]李兰,刘金朝,王成国.城轨车辆耐碰撞结构的数字设计研究[J].铁道机车车辆,2008,28(2):28-32.(Li Lan,Liu Jin-zhao,Wang Cheng-guo.Numerical study on crashworthy structure for urbanrail vellicle[J].Railway Locomotive&Car,2008,28(2):28-32.)
[2]Sun G,Li G,S Hou.Crashworthiness design for functionally graded foamfilled thin-walled structures[J].Materials Science&Engineering A,2010,527(7):1911-1919.
[3]罗昌杰,刘荣强,邓宗全.泡沫铝填充薄壁金属管塑性变形缓冲器吸能特性的试验研究[J].振动与冲击,2009,28(10):26-30.(Luo Chang-jie,Liu Rong-qiang,Deng Zong-quan.Experimental study on a plastic deformation energy absorber filled with aluminum foam in a thin-walled metal tube[J].Journal of Vibration and Shock,2009,28(10):26-30.)
[4]段尚国,赵洪伦.泡沫铝填充件对薄壁圆管结构压缩稳定性提高的研究[J].佳木斯大学学报:自然科学版,2011,29(6):801-805.(Duan Shang-guo,Zhao Hong-lun.Research on the effect of aluminum foam on the stability of vehicle passive safety device in crash[J].Journal of Jiamusi University:Natural Science Edition,2011,29(6):801-805.)
[5]Su chao Xie,Hui Zhou.Forecasting impact injuries of unrestrained occupants in railway vehicle passenger compartments[J].Traffic Injury Prevention,2014,15(7):740-7.
[6]武和全,辛勇.基于神经网络的变截面梁抗撞性分析及优化设计[J].振动与冲击,2010,29(10):102-107.(Wu He-quan,Xin Yong.Analysis and optimization design of beam with variable cross section based on neural network[J].Journal of Vibration and Shock,2010,29(10):102-107.)
[7]杨巨文,唐治,何峰.矿用扩径式吸能构件吸能防冲特性研究[J].振动与冲击,2015,34(8):134-138.(Yang Ju-wen,Tang Zhi,He Feng.Analysis on energy absorption and antiimpact properties of mine diameter expanding energy absorption components[J].Journal of Vibration and Shock,2015,34(8):134-138.)
[8]李小荣,郭永刚.基于遗传算法优化神经网络权值的损伤识别[J].噪声与振动控制,2008,28(6):47-51.(Li Xiao-rong,GuoYong-gang.Damage identif-acationon the base of optimic neural network weight by genetic algorithm[J].Noise and Vibration Control,2008,28(6):47-51.)
[9]张治国,张谢东.基于遗传算法和神经网络的结构损伤识别研究[J].水利与建筑工程学报,2005(3):1-3.(Zhang Zhi-guo,Zhang Xie-dong.Research on structural damage identification based on genetic algorithm and neural networks[J].Journal of Water Resources and Architectural Engineering,2005(3):1-3.)
[10]邹翔,高广军,董海鹏.高速列车多边形多胞吸能管耐撞性分析与优化[J].铁道科学与工程学报,2016,13(7):1386-1392.(Zou Xiang,Gao Guang-jun,Dong Hai-peng.Crashworthiness analysis and multi-objective optimization of multi-cell tube for high-speed train[J].Journal of Railway Science and Engineering,2016,13(7):1386-1392.)
[11]董喜文,康国政,朱志武.双层分级吸能结构优化设计[J].西华大学学报:自然科学版,2014,33(6):37-41.(Dong Xi-wen,Kang Guo-zheng,Zhu Zhi-wu.Optimization of boublelayer hierarchical energy absorbing structure[J].Journal of Xihua University:Natural Science Edition,2014,33(6):37-41.)
[12]Choi W M,J S Kim,H S Jung.Effect of punch angle on energy absorbing characteristics of tube-type crash elements[J].International Journal of Automotive Technology,2011,12(3):383-389.
[13]Choi W M,T S Kwon,H S Jung.Study on rupture of tube type crash energy absorber using finite element method[J].World Academy of Science Engineering&Technology,2013(76):595.
[2]Sun G,Li G,S Hou.Crashworthiness design for functionally graded foamfilled thin-walled structures[J].Materials Science&Engineering A,2010,527(7):1911-1919.
[3]罗昌杰,刘荣强,邓宗全.泡沫铝填充薄壁金属管塑性变形缓冲器吸能特性的试验研究[J].振动与冲击,2009,28(10):26-30.(Luo Chang-jie,Liu Rong-qiang,Deng Zong-quan.Experimental study on a plastic deformation energy absorber filled with aluminum foam in a thin-walled metal tube[J].Journal of Vibration and Shock,2009,28(10):26-30.)
[4]段尚国,赵洪伦.泡沫铝填充件对薄壁圆管结构压缩稳定性提高的研究[J].佳木斯大学学报:自然科学版,2011,29(6):801-805.(Duan Shang-guo,Zhao Hong-lun.Research on the effect of aluminum foam on the stability of vehicle passive safety device in crash[J].Journal of Jiamusi University:Natural Science Edition,2011,29(6):801-805.)
[5]Su chao Xie,Hui Zhou.Forecasting impact injuries of unrestrained occupants in railway vehicle passenger compartments[J].Traffic Injury Prevention,2014,15(7):740-7.
[6]武和全,辛勇.基于神经网络的变截面梁抗撞性分析及优化设计[J].振动与冲击,2010,29(10):102-107.(Wu He-quan,Xin Yong.Analysis and optimization design of beam with variable cross section based on neural network[J].Journal of Vibration and Shock,2010,29(10):102-107.)
[7]杨巨文,唐治,何峰.矿用扩径式吸能构件吸能防冲特性研究[J].振动与冲击,2015,34(8):134-138.(Yang Ju-wen,Tang Zhi,He Feng.Analysis on energy absorption and antiimpact properties of mine diameter expanding energy absorption components[J].Journal of Vibration and Shock,2015,34(8):134-138.)
[8]李小荣,郭永刚.基于遗传算法优化神经网络权值的损伤识别[J].噪声与振动控制,2008,28(6):47-51.(Li Xiao-rong,GuoYong-gang.Damage identif-acationon the base of optimic neural network weight by genetic algorithm[J].Noise and Vibration Control,2008,28(6):47-51.)
[9]张治国,张谢东.基于遗传算法和神经网络的结构损伤识别研究[J].水利与建筑工程学报,2005(3):1-3.(Zhang Zhi-guo,Zhang Xie-dong.Research on structural damage identification based on genetic algorithm and neural networks[J].Journal of Water Resources and Architectural Engineering,2005(3):1-3.)
[10]邹翔,高广军,董海鹏.高速列车多边形多胞吸能管耐撞性分析与优化[J].铁道科学与工程学报,2016,13(7):1386-1392.(Zou Xiang,Gao Guang-jun,Dong Hai-peng.Crashworthiness analysis and multi-objective optimization of multi-cell tube for high-speed train[J].Journal of Railway Science and Engineering,2016,13(7):1386-1392.)
[11]董喜文,康国政,朱志武.双层分级吸能结构优化设计[J].西华大学学报:自然科学版,2014,33(6):37-41.(Dong Xi-wen,Kang Guo-zheng,Zhu Zhi-wu.Optimization of boublelayer hierarchical energy absorbing structure[J].Journal of Xihua University:Natural Science Edition,2014,33(6):37-41.)
[12]Choi W M,J S Kim,H S Jung.Effect of punch angle on energy absorbing characteristics of tube-type crash elements[J].International Journal of Automotive Technology,2011,12(3):383-389.
[13]Choi W M,T S Kwon,H S Jung.Study on rupture of tube type crash energy absorber using finite element method[J].World Academy of Science Engineering&Technology,2013(76):595.