Probabilistic model and calibration of flexure strength of reinforced concrete columns
Received:June 11, 2019  Revised:August 07, 2019
View Full Text  View/Add Comment  Download reader
DOI:10.7511/jslx20190611001
KeyWord:reinforced concrete columns  flexure strength  uncertainties  probabilistic model  probabilistic calibration
           
AuthorInstitution
刘圣宾 广西大学 土木建筑工程学院, 南宁 ;南京市市政设计研究院有限责任公司, 南京
丁自豪 广西大学 土木建筑工程学院, 南宁
张永兵 广西大学 土木建筑工程学院, 南宁
余波 广西大学 土木建筑工程学院, 南宁 ;工程防灾与结构安全教育部重点实验室, 南宁 ;广西防灾减灾与工程安全重点实验室, 南宁
Hits: 692
Download times: 553
Abstract:
      A probabilistic model and a calibration method for flexure strength of reinforced concrete (RC) columns were proposed to overcome the disadvantage that traditional deterministic flexure strength models and calibration methods cannot rationally consider the influence of uncertainties.Based on the basic assumption of flexure strength analysis of RC members,deterministic flexure strength models for large (small) eccentric compression RC columns were established firstly according to the force balance condition of eccentric compression RC columns.Then analytical expressions for probabilistic flexure strength models of RC columns were developed by considering the influences of both aleatory and epistemic uncertainties.Subsequently,probabilistic models were established by determining the posterior distribution information of probabilistic model parameters based on the Bayesian theory and the Markov Chain Monte Carlo(MCMC)method.Finally,probabilistic calibration methods of traditional deterministic flexure strength models were proposed based on the probability density function,confidence interval and confidence level determined by the proposed probabilistic models.Analysis results show that the probabilistic flexure strength models not only can reasonably describe the probabilistic characteristics of flexure strength,but also provide an efficient way to calibrate the accuracy and confidence level of traditional deterministic flexure strength models of RC columns.