Bayesian Theory-Based Seismic Failure Modes Identification of Reinforced Concrete Columns

ABSTRACT

The aim of this research is to develop a probability method that can identify the flexure, shear or flexural-shear failure of RC columns. A total of 336 columns of cyclic load test results were collected, and a probabilistic model was constructed using the Bayesian procedure. Then, the model was simplified by Bayesian updating. The boundaries for each failure mode were selected through an internal search optimization. The results show that the Bayesian methodology can accurately predict the seismic failure modes of RC columns and the comparison of the identification with other methods underscore the validity of Bayesian method.

KEYWORDS:

• RC columns
• failure mode
• cyclic load
• bayesian theory
• probabilistic identification

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (Grant Nos. 51408223, 52008186, 51979109) and the Key Laboratory of Building Collapse Mechanism and Disaster Prevention, China Earthquake Administration (Grant No. FZ211102) and the Key Scientific and Technological Project of Henan Province (Grant No. 212102310935).