基于GARCH效应和改进罚指标的时域非线性损伤识别

doi:10.16183/j.cnki.jsjtu.2019.99.003

摘要/Abstract

摘要： 结构在服役期间可能会产生裂纹等损伤，裂纹的张开和闭合效应使损伤呈现出时域非线性特性.另外，基于时域模型的方法主要反映了刚度损伤引起相邻节点自由度处的特征改变，难以直接反映层间刚度的损伤信息.为了解决这些问题，提出了一种基于GARCH模型和改进罚指标的损伤识别方法.首先描述了GARCH模型的基本原理，并给出了GARCH模型的定阶方法和模型参数估计法.然后分析了时域非线性的双线性刚度特性，并提出了基于GARCH模型的非线性损伤识别原理和基于刚度的基本GARCH指标.最后利用罚函数原理，对层间刚度损伤提出了一种改进GARCH罚指标方法以提高识别可靠性.数值计算和实验研究结果表明，采用改进GARCH罚指标可以较好识别出结构非线性损伤，其识别效果要好于基本GARCH指标以及基于线性自回归模型和倒谱测距的方法.

关键词: 损伤识别； GARCH模型；罚函数；非线性；加速度响应

Abstract: In service period, some structures may exist damages like cracks, and the opening and closing of cracks make the damage display a time-domain nonlinear characteristic. In addition，time-domain model methods mainly provide the characteristic change of degree of freedom of adjacent nodes caused by stiffness damage, but it is difficult to directly show the inter-storey stiffness information. To solve these problems, a damage detection method based on GARCH model and improved penalty index was presented. Firstly, basic theory of GARCH model was described, and the order estimation and the parameter estimation of GARCH model were proposed. Then, the bilinear stiffness characteristic of time-domain nonlinearity was analyzed, and the nonlinear damage identification principle and basic GARCH index of stiffness identification were presented. Finally, an improved GARCH penalty index method was established to enhance the identification reliability for the inter-storey stiffness damage. Simulation and experiment results indicate that the improved GARCH penalty index can well identify the structural nonlinear damage, and the identification effect of the proposed index is obviously better than those of the basic GARCH index and those combined with autoregressive model and the cepstral metric index.

Key words: damage identification; GARCH model; penalty function; nonlinearity; acceleration response

中图分类号:

TB 123

郭惠勇，黄淇. 基于GARCH效应和改进罚指标的时域非线性损伤识别[J]. 上海交通大学学报, 2019, 53(11): 1326-1334.

GUO Huiyong，HUANG Qi. Time-Domain Nonlinear Damage Detection Based on GARCH Effect and Improved Penalty Index[J]. Journal of Shanghai Jiaotong University, 2019, 53(11): 1326-1334.

参考文献

［1］LI P J, XU D W, ZHANG J. Probability-based structural health monitoring through Markov chain Monte Carlo sampling［J］. International Journal of Structural Stability and Dynamics, 2016, 16(7): 1550039. ［2］BAO C X, HAO H, LI Z X. Integrated ARMA model method for damage detection of subsea pipeline system［J］. Engineering Structures, 2013, 48: 176-192. ［3］李万润, 杜永峰, 倪一清, 等. 基于伪传递函数的高耸结构损伤识别［J］. 振动、测试与诊断, 2015, 35(1): 63-69. LI Wanrun, DU Yongfeng, NI Yiqing, et al. High-rise structure damage identification based on pseudo-transfer function ［J］. Journal of Vibration，Measurement & Diagnosis, 2015, 35(1): 63-69. ［4］FASSOIS S D, SAKELLARIOU J S. Time-series methods for fault detection and identification in vibrating structures［J］. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2007, 365(1851): 411-448. ［5］ZHENG H T, MITA A. Damage indicator defined as the distance between ARMA models for structural health monitoring［J］. Structural Control & Health Monitoring, 2008, 15(7): 992-1005. ［6］DA SILVA S, DIAS M, LOPES V. Damage detection in a benchmark structure using AR-ARX models and statistical pattern recognition［J］. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2007, 29(2): 174-184. ［7］NAIR K K, KIREMIDJIAN A S, LAW K H. Time series-based damage detection and localization algorithm with application to the ASCE benchmark structure［J］. Journal of Sound and Vibration, 2006, 291(1/2): 349-368. ［8］SAAED T E，NIKOLAKOPOULOS G. Identification of building damage using ARMAX model: A parametric study［J］. Diagnostyka, 2016, 17(3): 3-14. ［9］罗德河, 郑东健, 甘声玄, 等. 基于ARMAX模型的混凝土坝损伤诊断［J］. 武汉大学学报(工学版), 2018, 51(4): 294-298. LUO Dehe, ZHENG Dongjian, GAN Shengxuan, et al. Damage diagnosis of concrete dams based on ARMAX model［J］. Engineering Journal of Wuhan University, 2018, 51(4): 294-298. ［10］BOLLERSLEV T. Generalized autoregressive conditional heteroskedasticity［J］. Journal of Econometrics, 1986, 31(3): 307-327. ［11］KIM Y, HWANG E. A dynamic Markov regime-switching GARCH model and its cumulative impulse response function［J］. Statistics & Probability Letters, 2018, 139: 20-30. ［12］ZHANG Y J, ZHANG Y X, DENG Z M, et al. Sea surface target detection based on complex ARMA-GARCH processes［J］. Digital Signal Processing, 2017, 70: 1-13. ［13］CHEN L J, YU L. Structural nonlinear damage identification algorithm based on time series ARMA/GARCH model［J］. Advances in Structural Engineering, 2013, 16(9): 1597-1609. ［14］JIANG W, RUAN Q S, LI J F, et al. Modeling returns volatility: Realized GARCH incorporating realized risk measure［J］. Physica A: Statistical Mechanics and Its Applications, 2018, 500: 249-258. ［15］XING Z H, MITA A. Locating the damaged storey of a building using distance measures of low-order AR models［J］. Smart Structures and Systems, 2010, 6(9): 991-1005. ［16］赵仕通, 肖黎, 屈文忠. 框架结构非线性损伤的主成分分析识别方法研究［J］. 机械科学与技术, 2018, 37(1): 8-12. ZHAO Shitong, XIAO Li, QU Wenzhong. Study on nonlinear damage identification of frame structure with a principal component analysis method［J］. Mechanical Science and Technology for Aerospace Engineering, 2018, 37(1): 8-12.

[1]	陈炉云，杨念. 基于多层次权重系数的结构动力学模型修正[J]. 上海交通大学学报（自然版）, 2017, 51(12): 1415-1421.
[2]	包继虎a, 张鹏a, b, 朱昌明b. 变长度柔性提升系统钢丝绳横向振动建模及分析[J]. 上海交通大学学报（自然版）, 2012, 46(03): 341-345.
[3]	陶建峰，马琼，商元，王旭永. 双马达驱动弹性圆柱负载的动力学分析[J]. 上海交通大学学报（自然版）, 2008, 42(11): 1892-1895.
[4]	杨坤，梅志远，李华东. 正交加筋复合材料夹层板的自由振动求解[J]. 上海交通大学学报（自然版）, 2014, 48(06): 863-869.
[5]	李明，李富才，李鸿光，孟光. 浅裂纹转子特性的动态圣维南原理解释[J]. 上海交通大学学报（自然版）, 2014, 48(06): 870-876.