吴来义,戴新军,贾文文

• 1:中铁大桥(南京)桥隧诊治有限公司

摘要(Abstract)：

针对桥梁健康监测系统支座位移预警阈值的设置问题,该文采用相空间重构与滑动窗口Elman神经网络预测模型相结合的方法进行研究。以国内某特大跨径拱桥支座位移作为研究对象,选用改进的C-C算法确定嵌入维数和延迟时间,通过对一维时间序列的重构来恢复系统原动力特征。同时,为避免Elman神经网络训练时长过大、不能在线及时更新参数的问题,采用滑动窗口方法对Elman神经网络训练数据进行实时更新,建立基于相空间重构的滑动窗口Elman神经网络预警模型,对支座位移预测均值进行区间估计。实验表明该预警方法能够得到区间型的位移阈值,可以敏感地发现位移的异常变化,实现多级预警,确保桥梁安全。

关键词(KeyWords)： 相空间重构;滑动窗口;Elman神经网络;多级预警

Abstract:

Keywords:

基金项目(Foundation): 江苏省科技厅重点研发计划项目(BE2018049);; 国家重点研发计划项目(2016YEC0802202,2017YFC0806001);; 江苏省科技计划项目(BZ2017056)

作者(Author): 吴来义,戴新军,贾文文

DOI: 10.16251/j.cnki.1009-2307.2020.04.007

参考文献(References)：

• [1] DING Y L,SUN P,WANG G X,et al.Early-warning method of train running safety of a high-speed railway bridge based on transverse vibration monitoring[J/OL].[2019-09-24].http://connection.ebscohost.com/c/articles/109249779/early-warning-method-train-running-safety-high-speed-railway-bridge-based-transverse- vibration-monitoring.
• [2] ZHONG J M,SEN S L.Interval threshold setting method of vertical support reaction for early warning of concrete bridges[J].Applied Mechanics and Materials,2014 (638/640):1049-1054.
• [3] 赵瀚玮,丁幼亮,李爱群,等.大跨多线高速铁路钢桁拱桥车:桥振动安全预警研究[J].中国铁道科学,2018,39(2):28-36.(ZHAO Hanwei,DING Youliang,LI Aiqun,et al.Research on safety early warning of vehicle-bridge vibration for long-span multi-track steel-truss arch bridge of high speed railway[J].China Railway Science,2018,39(2):28-36.)
• [4] 刘小玲,黄侨,任远,等.大跨度钢斜拉桥主梁监测挠度的评估与预警[J].湖南大学学报(自然科学版),2016,43(9):98-104.(LIU Xiaoling,HUANG Qiao,REN Yuan,et al.Assessment and early warning on the monitoring girder deflection of the long-span steel cable-stayed bridge [J].Journal of Hunan University(Natural Science),2016,43(9):98-104.)
• [5] 苏成,廖威,袁昆,等.桥梁健康监测在线预警指标研究[J].桥梁建设,2015,45(3):44-50.(SU Cheng,LIAO Wei,YUAN Kun,,et al.Research of online early warning indices for bridge health monitoring [J].Bridge Construction,2015,45(3):44-50.)
• [6] 黄海宾.基于监测数据相关性建模的大跨桥梁预警方法研究[D].大连:大连理工大学,2017.(HUANG Haibin.Early-warning methods for long-span bridges based on monitoring date correlation modeling [D].Dalian:Dalian University of Technology,2017.)
• [7] 胡铁明,苟红兵,张冠华,等.基于温度与支座位移相关性的斜拉桥损伤预警[J].沈阳大学学报(自然科学版),2015,27(1):55-59.(HU Tieming,GOU Hongbing,ZHANG Guanhua,et al.Damage alarming for cable-stayed bridge based on correlation of temperature and displacement [J].Journal of Shenyang University(Natural Science),2015,27(1):55-59.)
• [8] PACKARD N H,CRUTCHIFIELD J P,FARMER J D,et al.Geometry from a time series[EB/OL].[2019-09-24].https://www.scirp.org/reference/ReferencesPapers.aspx?ReferenceID=2377866.
• [9] TAKENS F.Detecting strange attractors in fluid turbulence[A].[2019-09-24].https://link.springer.com/chapter/10.1007/BFb0091924.
• [10] JORDAN M I.Attractor dynamics and parallelism in a connectionist sequential machine[M].NJ,USA:,IEEE Press,1986,:531-546.
• [11] ELMAN J L.Finding Structure in time[J].Cognitive Science,1990,14(2):179-211.
• [12] 陆振波,蔡志明,姜可宇.基于改进的C-C方法的相空间重构参数选择[J].系统仿真学报,2007,19(11):2527-2529,2538.(LU Zhenbo,CAI Zhiming,JIANG Keyu.Determination of embedding parameters for phase space reconstruction based on improved C-C method [J].Journal of System Simulation,2007,19(11):2527-2529,2538.)
• [13] ALEX A.Dynamical recurrent neural networks towards prediction and modeling of dynamical systems[J].Neurocom Puting,1999,28(1/3):207.