TY - GEN
T1 - Finite element model updating using the shadow Hybrid Monte Carlo technique
AU - Boulkaibet, I.
AU - Mthembu, L.
AU - Marwala, T.
AU - Friswell, M. I.
AU - Adhikari, S.
PY - 2013
Y1 - 2013
N2 - Recent research in the field of Finite Element Model (FEM) updating has advocated the benefits of adopting Bayesian analysis techniques. These techniques are well suited to dealing with the uncertainties associated with complex systems. However, Bayesian formulations require the evaluation of the Posterior Distribution Function (pdf) which may not be available in analytical form. This is the case in FEM updating. In such cases sampling methods can provide good approximations of the Posterior distribution when implemented in the Bayesian context. In this paper, we propose the use of the Shadow Hybrid Monte Carlo (SHMC) technique for the problem of determining the most probable FEM updating parameters for the given data. SHMC is based on Hybrid Monte Carlo (HMC) and designed to improve sampling by allowing for larger system sizes and time steps. The accuracy and efficiency of this sampling method is tested on the updating of a structural beam models.
AB - Recent research in the field of Finite Element Model (FEM) updating has advocated the benefits of adopting Bayesian analysis techniques. These techniques are well suited to dealing with the uncertainties associated with complex systems. However, Bayesian formulations require the evaluation of the Posterior Distribution Function (pdf) which may not be available in analytical form. This is the case in FEM updating. In such cases sampling methods can provide good approximations of the Posterior distribution when implemented in the Bayesian context. In this paper, we propose the use of the Shadow Hybrid Monte Carlo (SHMC) technique for the problem of determining the most probable FEM updating parameters for the given data. SHMC is based on Hybrid Monte Carlo (HMC) and designed to improve sampling by allowing for larger system sizes and time steps. The accuracy and efficiency of this sampling method is tested on the updating of a structural beam models.
KW - Bayesian
KW - Finite element model updating
KW - Hybrid monte carlo method
KW - Markov chain monte carlo
KW - Sampling
KW - Shadow hybrid monte carlo
UR - http://www.scopus.com/inward/record.url?scp=84889594992&partnerID=8YFLogxK
U2 - 10.1007/978-1-4614-6546-1_53
DO - 10.1007/978-1-4614-6546-1_53
M3 - Conference contribution
AN - SCOPUS:84889594992
SN - 9781461465454
T3 - Conference Proceedings of the Society for Experimental Mechanics Series
SP - 489
EP - 498
BT - Special Topics in Structural Dynamics - Proceedings of the 31st IMAC, A Conference on Structural Dynamics, 2013
T2 - 31st International Modal Analysis Conference on Structural Dynamics, IMAC 2013
Y2 - 11 February 2013 through 14 February 2013
ER -