[1]浦炜钦,宋志刚,胡一鸣,等.基于BP神经网络的单人步行荷载参数识别[J].地震工程与工程振动,2022,42(02):193-200.[doi:10.13197/j.eeed.2022.0219]
　PU Weiqin,SONG Zhigang,HU Yiming,et al.Single pedestrian walking load identification based on BP neural network[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2022,42(02):193-200.[doi:10.13197/j.eeed.2022.0219]
点击复制

基于BP神经网络的单人步行荷载参数识别
分享到：

《地震工程与工程振动》[ISSN:/CN:]

卷:
42
期数:
2022年02期
页码:
193-200
栏目:
论文
出版日期:
2022-04-30

文章信息/Info

Title:
Single pedestrian walking load identification based on BP neural network
作者:
浦炜钦;  宋志刚;  胡一鸣;  张博皓
昆明理工大学 建筑工程学院, 云南 昆明 650500
Author(s):
PU Weiqin;  SONG Zhigang;  HU Yiming;  ZHANG Bohao
Faculty of Civil Engineering and Architecture, Kunming University of Science and Technology, Kunming 650500, China
关键词:
步行荷载; 参数识别; 均匀试验设计; BP神经网络
Keywords:
walking load; parameter identification; uniform design; BP neural network
分类号:
TU312+.1
DOI:
10.13197/j.eeed.2022.0219
摘要:
步行荷载模型的参数识别是准确重构步行荷载的前提。通过试验得到单人不同步频下的步行荷载加速度时程，经过傅里叶变换得到步行荷载频谱，获得各阶动载因子与实际步频。基于包含亚谐分量的步行荷载傅里叶级数模型，提出均匀试验设计与BP神经网络相结合的相位角识别方法，进而重构步行荷载。结果表明，提高均匀试验设计的水平数可以提高相位角的识别精度与重构步行荷载的准确性，各步频实测－重构步行荷载的均方根相对误差最小值约为9%，最大值约为21%，均值约为15%，在重构步行荷载时考虑亚谐分量可以更好地反映行人左右脚的步行荷载差异，步行荷载主、亚谐分量的各阶相位角大致服从变异系数为30%～70%的正态分布。
Abstract:
The parameter identification of walking load model is the premise of reconstructing walking load.Through the experiments,we obtain the walking load time history of single pedestrian at different walking frequencies. Through Fourier transform,we obtain the walking load spectrum,actual walking frequency and dynamic load factor. Uniform design and BP neural network were built to identify phase angle of walking load based on the Fourier series model of walking load with sub-harmonic components.Then,the correctness of the proposed method was verified by comparing the RMSE,load time history and load spectrum of measured-reconstructed walking load.Results show that the improvement of the level of uniform design can improve the accuracy of phase angle and reconstructed walking load,the minimum RMSE of measured-reconstructed walking load is about 9%,the maximum value is about 21%,the mean value is about 15%,the reconstructed walking load with sub-harmonic components can reflect the difference of walking load between left and right feet,the phase angles of the harmonic and sub-harmonic components obey normal distribution with the variation coefficient of 30%~70%.

参考文献/References:

[1] BROWNJOHN J M W,PAVIC A,OMENZETTER P. A spectral density approach for modelling continuous vertical forces on pedestrian structures due to walking[J]. Canadian Journal of Civil Engineering,2004,31(4):65-77.
[2] SAHNACI C,KASPERSKI M. Random loads induced by walking[C]//Proceedings of the 6th European Conference on Structural Dynamics. Rotterdam:Millpress,2005:441-446.
[3] ELLIS B R,JI T. Floor vibration induced by dance-type loads:Theory[J]. Structural Engineer,1994,72(3):37-44.
[4] IVANOVI?S ?PAVIC A,REYNOLDS P. Probability-based prediction of multi-mode vibration response to walking excitation[J]. Engineering Structures,2007,29(6):942-954.
[5] 操礼林. 高铁候车厅大跨楼盖人致振动回应分析与减振控制研究[D]. 南京:东南大学,2016. CAO Lilin. Research on Human-induced Vibration Response and Vibration Control of Long-span Floor in High-speed Railway Waiting Hall[D].Nanjing:Southeast University,2016.(in Chinese)
[6] 陈隽,王浩祺,彭怡欣. 行走激励的傅里叶级数模型及其参数的实验研究[J]. 振动与冲击,2014,33(8):11-15. CHEN Jun,WANG Haoqi,PENG Yixin. Experimental investigation on Fourier series model of walking load and its coefficients[J]. Journal of Vibration and Shock,2014,33(8):11-15.(in Chinese)
[7] CHEN Jun,DING Guo,STANA ZIVANOVIC. Stochastic single footfall trace model for pedestrian walking load[J]. International Journal of Structural Stability and Dynamics,2019,19(3):1950029.
[8] 谢伟平,章涛,何卫,等. 单人步行荷载模型研究[J]. 振动与冲击,2018,37(14):214-220. XIE Weiping,ZHANG Tao,HE Wei,et al. Load model for a single pedestrian[J]. Journal of Vibration and Shock,2018,37(14):214-220.(in Chinese)
[9] 王琪,宋志刚,王子洋. 采用加速度传感器测定竖向步行力的试验研究[J]. 振动与冲击,2019,38(1):215-220. WANG Qi,SONG Zhigang,WANG Ziyang. Tests for measuring vertical pedestrian loads using acceleration sensors[J]. Journal of Vibration and Shock,2019,38(1):215-220.(in Chinese)
[10] 史峰,王小川,郁磊,等. MATLAB神经网络30个案例分析[M]. 北京:北京航空航天大学出版社,2010:11-36. SHI Feng,WANG Xiaochuan,YU Lei,et al. MATLAB Neural Network 30 Case Analysis[M]. Beijing:Beijing University of Aeronautics and Astronautics Press,2010:11-36.(in Chinese)
[11] SILVA D,BRITO F,PIMENTEL L. Modeling of crowd load in vertical direction using biodynamic model for pedestrians crossing footbridges[J]. Canadian Journal of Civil Engineering,2013,40(12):1196-1204.
[12] 方开泰. 均匀设计-数论方法在试验设计的应用[J]. 应用数学学报,1980,3(4):363-372. FANG Kaitai. Uniform design-application of number theory method in experimental design[J]. Acta Mathematicae Applicatae Sinica,1980,3(4):363-372.(in Chinese)
[13] 王元,方开泰. 关于均匀分布与试验设计(数论方法)[J]. 科学通报,1981(2):65-70. WANG Yuan,FANG Kaitai. About uniform distribution and experimental design(number theory method)[J]. Chinese Science Bulletin,1981(2):65-70.(in Chinese)
[14] 方开泰. 均匀设计及其应用[J]. 数理统计与管理,1994,13(1):57-63. FANG Kaitai. Uniform design and its application[J]. Journal of Applied Statistics and Management,1994,13(1):57-63.(in Chinese)

相似文献/References:

[1]易伟建,周云.基于高阶局部模态的弹性地基上框架结构物理参数识别研究[J].地震工程与工程振动,2007,27(01):117.
　Yi Weijian,Zhou Yun.Research on physical parameter identification of concrete frame structure on elastic foundation based on higher modes[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2007,27(02):117.
[2]瞿伟廉,刘嘉,涂建维,等.500kN足尺磁流变液阻尼器设计的关键技术[J].地震工程与工程振动,2007,27(02):124.
　QU Weilian,LIU Jia,TU Jianwei,et al.Crucial techniques for design of 500 kN large-scale MR damper[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2007,27(02):124.
[3]常军,孙利民,张启伟.随机子空间识别方法计算效率的改进[J].地震工程与工程振动,2007,27(03):088.
　CHANG Jun,SUN Limin,ZHANG Qiwei.Improvement in stochastic subspace identification[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2007,27(02):088.
[4]王祥建,崔杰.一种迭代算法在结构物理参数识别中的应用[J].地震工程与工程振动,2008,28(01):044.
　WANG Xiangjian,CUI Jie.An iterative algorithm applied to structural physical parameter identification[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2008,28(02):044.
[5]常军,孙利民,张启伟.基于两阶段稳定图的随机子空间识别结构模态参数[J].地震工程与工程振动,2008,28(03):047.
　CHANG Jun,SUN Limin,ZHANG Qiwei.Study on the method for stochastic subspace identifying structural modal parameters based on two-stage stabilization diagram[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2008,28(02):047.
[6]禹见达,陈政清,王修勇,等.基于遗传算法的磁流变阻尼器模型参数识别[J].地震工程与工程振动,2009,29(01):135.
　YU Jianda,CHEN Zhengqing,WANG Xiuyong,et al.Parameter identification of magnetorheological dampers based on genetic algorithm[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2009,29(02):135.
[7]陈关宝,于德介.基于多尺度线调频基稀疏信号分解的线性时变SDOF系统参数识别[J].地震工程与工程振动,2013,33(04):100.
　CHEN Guanbao,YU Dejie.The parameter identification of linear time-varying SDOF systems based on multi-scale chirplet sparse signal decomposition[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2013,33(02):100.
[8]徐丽,易伟建,吴高烈.混凝土框架模型结构参数的识别[J].地震工程与工程振动,2006,26(04):121.
　Xu Li,Yi Weijian,Wu Gaolie.Parameter identification of a model concrete frame structure[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2006,26(02):121.
[9]常军,张启伟,孙利民.随机子空间方法在桥塔模态参数识别中的应用[J].地震工程与工程振动,2006,26(05):183.
　Chang Jun,Zhang Qiwei,Sun Limin.Application of stochastic subspace identification in modal parameter identification of bridge tower[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2006,26(02):183.
[10]夏遵平,王彤,张永年.基于Hilbert谱变换函数的运行模态分析方法[J].地震工程与工程振动,2014,34(06):097.[doi:10.13197/j.eeev.2014.06.97.xiazp.013]
　XIA Zunping,WANG Tong,ZHANG Yongnian.Operational modal analysis based on Hilbert transformation[J].EARTHQUAKE ENGINEERING AND ENGINEERING DYNAMICS,2014,34(02):097.[doi:10.13197/j.eeev.2014.06.97.xiazp.013]

备注/Memo

备注/Memo:
收稿日期:2021-4-30;改回日期:2021-8-11。
基金项目:云南省科技厅重点研发计划项目（KKS7202006003）
作者简介:浦炜钦(1996-),男,硕士研究生,主要从事人致振动研究.E-mail:1770672328@qq.com
通讯作者:宋志刚(1974-),男,教授,博士,主要从事工程振动、火灾风险防控和混凝土研究.E-mail:80944459@qq.com
更新日期/Last Update: 1900-01-01