The 2nd ZHITU Symposium on Advances in Civil Engineering
Abstract
This study proposes a probabilistic method to predict the time-history deflections of railway bridges subject. The Gaussian process regression, a non-parametric regression method, is introduced to construct the probabilistic prediction model where the inherent uncertainty of time-history deflection data is considered. The applicability of the proposed method is discussed with the actual measurement data of a railway bridge in South Korea.
Publisher
Ulsan National Institute of Science and Technology