【摘要】 收集国内水工钢闸门锈蚀检测的数据,采用概率统计方法和3σ方法确定了现役钢闸门锈蚀速率的概率分布模型.提出基于贝叶斯理论的钢闸门锈蚀速率更新的方法,并采用算例说明了所提方法的有效性.结果表明:我国水工钢闸门锈蚀速率的均值和变异系数分别为0.034 mm/a和0.43,锈蚀速率均值的均值和变异系数分别为0.034 mm/a和0.60,它和国外普通碳钢在淡水中的锈蚀速率大体相当.锈蚀速率不拒绝正态分布和对数正态分布,但以对数正态分布为优.采用贝叶斯理论更新后的锈蚀速率的不确定性明显减小.更多还原

【Abstract】 Determining corrosion rate of existing hydraulic steel gates is essential to evaluate the safety and predict the service life of hydraulic gates.Therefore this paper aims to propose a methodology to determine the probabilistic model of corrosion rate for existing hydraulic steel gates.Based on the collected corrosion data of hydraulic gates from on-site inspections in China,the statistical approach along with 3σ approach is adopted to obtain probabilistic model of corrosion rate.The updating of corrosion rate based on the inspected corrosion data in a specific hydraulic steel gate is formulated in the Bayesian framework.The results indicate that the mean and coefficient of variation of corrosion rate are 0.034(mm/a) and 0.43,respectively.The mean and coefficient of variation of mean corrosion rate for hydraulic steel gates are 0.034(mm/a) and 0.60,respectively,which is approximately equal to that under fresh water condition in abroad.Through the K-S goodness-of-fit test at the 5% significance level,both normal and lognormal distributions for corrosion rate can not be rejected,but the lognormal distribution model is better than the normal distribution.The uncertainties in corrosion rate are reduced substantially after being updated with the information from inspections of hydraulic gates.更多还原