摘要
针对我国应用较多的美国桥梁设计规范中可靠度指标的计算公式,从概率分布和变异系数两方面对其进行误差原因分析,指出无论荷载效应是否服从对数正态分布,该公式均是近似计算公式。用该公式和蒙特卡罗法分别对工程中常见的8种工况进行计算,结果表明:该公式的误差一般在30%左右,最高可达46%,计算得到的可靠指标均小于蒙特卡罗法的计算结果。最后从理论和数值分析2个角度对该公式进行修正,并用修正后的公式计算了8种工况的可靠指标。计算结果表明,计算精度明显提高,误差一般能控制在11%以内。
A formula for calculating the reliability index in the American Bridges Design Specifications is widely applied in China. Its error cause is analyzed in terms of probability distribution and variation coefficient. It is pointed out that the formula is an approximate formula no matter whether the load effect follows lognormal distribution. The reliability indices are calculated by using this formula and the Monte Carlo Method under eight typical conditions. The results show that the error of the formula is normally around 30% and up to 46%, and the reliability indices obtained by the formula are always less than those by Monte Carlo Method. Finally, both methods of theoretical analysis and numerical analysis are put forward to revise the formula. The reliability indices under eight conditions are calculated with the revised formula. Results indicate that the calculation accuracy has been obviously improved and the error is usually controlled less than 11%.
出处
《中国铁道科学》
EI
CAS
CSCD
北大核心
2009年第4期41-45,共5页
China Railway Science
关键词
桥梁
可靠度指标
计算公式
蒙特卡罗法
理论修正
数值修正
Bridge
Reliability index
Calculation formula
Monte Carlo Method
Theoretic correction
Numerical correction