Existing studies on sparse Bayesian learning for structural damage detection usually assume that the posterior probability density functions follow standard distributions which facilitate to circumvent the intractable integration problem of the evidence by means of numerical sampling or analytical derivation. Moreover, the uncertainties of each mode are usually quantified as a common parameter to simplify the calculation. These assumptions may not be realistic in practice. This study proposes a sparse Bayesian method for structural damage detection suitable for standard and nonstandard probability distributions. The uncertainty corresponding to each mode is assumed as different. Variational Bayesian inference is developed and the posterior probability density functions of each unknown are individually derived. The parameters are found to follow the gamma distribution, whereas the distribution of the damage index cannot be directly obtained because of the nonlinear relationship in its posterior probability density function. The delayed rejection adaptive Metropolis algorithm is then adopted to generate numerical samples of the damage index. The coupled damage index and parameters in the variational Bayesian inference are successively calculated via an iterative process. A laboratory-tested frame is utilised to verify the effectiveness of the proposed method. The results indicate that the sparse damage can be accurately detected. The proposed method has the advantage of high accuracy and broad applicability.
ZhangJXuYLXiaY, et al. A new statistical moment-based structural damage detection method. Struct Eng Mech2008; 30: 445–466.
2.
SimoenEDe RoeckGLombaertG. Dealing with uncertainty in model updating for damage assessment: a review. Mech Syst Signal Pr2015; 56–57: 123–149.
3.
XiaYHaoH. Statistical damage identification of structures with frequency changes. J Sound Vib2003; 263: 853–870.
4.
DoeblingSWFarrarCRPrimeMB, et al. Damage identification and health monitoring of structural and mechanical systems from changes in their vibration characteristics: a literature review. Los Alamos National Laboratory Report, No. LA-13070-MS, 1996.
5.
SohnHFarrarCRHemezFM, et al. A review of structural health monitoring literature: 1996–2001. Los Alamos National Laboratory Report, No. LA-13976-MS, 2003.
6.
HuaXGNiYQChenZQ, et al. An improved perturbation method for stochastic finite element model updating. Int J Numerical Meth Eng2008; 73: 1845–1864.
7.
YeoIShinSLeeHS, et al. Statistical damage assessment of framed structures from static responses. J Eng Mech ASCE2000; 126: 414–421.
8.
GulMCatbasFN. Statistical pattern recognition for structural health monitoring using time series modeling: theory and experimental verifications. Mech Syst Signal Pr2009; 23: 2192–2204.
9.
BeckJLKatafygiotisLS. Updating models and their uncertainties. I: Bayesian statistical framework. J Eng Mech ASCE1998; 124: 455–461.
10.
ChingJBeckJL. Bayesian analysis of the phase II IASC-ASCE structural health monitoring experimental benchmark data. J Eng Mech ASCE2004; 130: 1233–1244.
11.
ZhuYCAuSK. Bayesian operational modal analysis with asynchronous data, part I: most probable value. Mech Syst Signal Pr2018; 98: 652–666.
12.
YanWJKatafygiotisLS. An analytical investigation into the propagation properties of uncertainty in a two-stage fast Bayesian spectral density approach for ambient modal analysis. Mech Syst Signal Pr2019; 118: 503–533.
13.
HuangYShaoCSWuBA, et al. State-of-the-art review on Bayesian inference in structural system identification and damage assessment. Adv Struct Eng2019; 22: 1329–1351.
14.
TippingME. Sparse Bayesian learning and the relevance vector machine. J Mach Learn Res2001; 1: 211–244.
15.
MalioutovDCetinMWillskyAS. A sparse signal reconstruction perspective for source localization with sensor arrays. IEEE T Signal Process2005; 53: 3010–3022.
16.
FangJShenYNLiFW, et al. Support knowledge-aided sparse Bayesian learning for compressed sensing. IEEE Int Conf Acoust2015: 3786–3790
17.
HuangYBeckJL. Hierarchical sparse Bayesian learning for structural health monitoring with incomplete modal data. Int J Uncertain Quan2015; 5: 139–169.
18.
HuangYBeckJLLiH. Hierarchical sparse Bayesian learning for structural damage detection: theory, computation and application. Struct Saf2017; 64: 37–53.
19.
HuangYBeckJLLiH. Bayesian system identification based on hierarchical sparse Bayesian learning and Gibbs sampling with application to structural damage assessment. Comput Method Appl M2017; 318: 382–411.
20.
HouRRXiaYZhouXQ, et al. Sparse Bayesian learning for structural damage detection using expectation-maximization technique. Struct Control Health2019; 26: e2343.
21.
WangXYHouRRXiaY, et al. Laplace approximation in sparse Bayesian learning for structural damage detection. Mech Syst Signal Pr2020; 140: 106701.
22.
BeckJL. Bayesian system identification based on probability logic. Struct Control Health2010; 17: 825–847.
23.
BealMJ. Variational algorithms for approximate Bayesian inference. PhD thesis, University of London, 2003.
24.
SmithAFMRobertsGO. Bayesian computation via the Gibbs sampler and related Markov-chain Monte-Carlo methods. J Roy Stat Soc B Met1993; 55: 3–23.
25.
BishopCM. Pattern recognition and machine learning. Berlin: Springer, 2006.
26.
ParisiG. Statistical field theory. Boston, MS: Addison-Wesley, 1988.
27.
BaldacchinoTCrossEJWordenK, et al. Variational Bayesian mixture of experts models and sensitivity analysis for nonlinear dynamical systems. Mech Syst Signal Pr2016; 66–67: 178–200.
28.
JacobsWRBaldacchinoTDoddT, et al. Sparse Bayesian nonlinear system identification using variational inference. IEEE T Automat Contr2018; 63: 4172–4187.
29.
BleiDMKucukelbirAMcAuliffeJD. Variational Inference: a review for statisticians. J Am Stat Assoc2017; 112: 859–877.
30.
AndrieuCde FreitasNDoucetA, et al. An introduction to MCMC for machine learning. Mach Lear2003; 50: 5–43.
31.
ChibSGreenbergE. Understanding the Metropolis-Hastings algorithm. Am St1995; 49: 327–335.
32.
ArmingerG. A Bayesian approach to nonlinear latent variable models using the Gibbs sampler and the Metropolis-Hastings algorithm. Psychometrika1998; 63: 271–300.
33.
AuSKBeckJL. Estimation of small failure probabilities in high dimensions by subset simulation. Prob Eng Mech2001; 16: 263–277.
34.
YangJHLamHF. An efficient adaptive sequential Monte Carlo method for Bayesian model updating and damage detection. Struct Control Health2018; 25: e2260.
35.
ChingJYChenYC. Transitional Markov chain Monte Carlo method for Bayesian model updating, model class selection, and model averaging. J Eng Mech2007; 133: 816–832.
36.
HaarioHSaksmanETamminenJ. Adaptive proposal distribution for random walk Metropolis algorithm. Computation Stat1999; 14: 375–395.
37.
HaarioHSaksmanETamminenJ. An adaptive Metropolis algorithm, Bernoulli2001; 7: 223–242.
38.
TierneyLMiraA. Some adaptive Monte Carlo methods for Bayesian inference. Stat Med1999; 18: 2507–2515.
39.
GreenPJMiraA. Delayed rejection in reversible jump Metropolis-Hastings. Biometrika2001; 88: 1035–1053.
40.
MiraA. On Metropolis-Hastings algorithms with delayed rejection. Metron2001; 59: 231–241.
41.
HaarioHLaineMMiraA, et al. DRAM: efficient adaptive MCMC. Stat Comput2006; 16: 339–354.
42.
ZuevKMKatafygiotisLS. Modified Metropolis-Hastings algorithm with delayed rejection. Prob Eng Mech2011; 26: 405–412.
43.
ZhangJWanCFSatoT. Advanced Markov Chain Monte Carlo approach for finite element calibration under uncertainty. Comput Aid Civ Inf2013; 28: 522–530.
44.
WanHPRenWX. Stochastic model updating utilizing Bayesian approach and Gaussian process model. Mech Syst Signal Pr2016; 70–71: 245–268.
45.
ZhouXQXiaYWengS. L1 regularization approach to structural damage detection using frequency data. Struct Health Monit2015; 14: 571–582.
46.
HouRRXiaYZhouXQ. Structural damage detection based on l1 regularization using natural frequencies and mode shapes. Struct Control Health2018; 25: e2107.
47.
ChenZPPanCDYuL. Structural damage detection via adaptive dictionary learning and sparse representation of measured acceleration responses. Measurement2018; 128: 377–387.
48.
WangLLuZR. Sensitivity-free damage identification based on incomplete modal data, sparse regularization and alternating minimization approach. Mech Syst Signal Pr2019; 120: 43–68.
49.
FormentiDRichardsonM. Parameter estimation from frequency response measurements using rational fraction polynomials (twenty years of progress). Proc Int Modal Anal Conf2002; 4753: 373–382.
