Frequency Identification of a historical Masonry Building Based on the Bayesian Inference

The frequency of historical masonry buildings is difficult to evaluate because of the uncertain nature of the material properties, existing damages and structural configurations. A Markov Chain Monte Carlo (MCMC) method based on the Bayesian Inference was proposed to deal with the issue. In the frequency identification for a historical masonry building, a prior probability density function (PDF) of the frequency was given based on subjective judgement and experience. The prior PDF was then updated based on measured frequency data. The updated PDF, or the posterior PDF, was used to deterimined the reasonable estimator for the 1st natural frequency. It was shown that the proposed method was applicableandrobust.