高强钢筋疲劳试验小子样空间的Bootstrap分析方法
The Bootstrap analysis method of high strength steel fatigue tests under small sample space

Bootstrap方法是一种利用再生样本来估计未知统计量估计值分布的方法。通过Bootstrap法的再抽样方法,可以拟合不同可靠度下的P-S-N曲线,同时给出其相应的置信区间。以高速铁路用HRB400高强钢筋闪光对焊接头的疲劳试样数据为例,运用t分布调整法、ASTM法以及Bootstrap方法,计算高强钢筋P-S-N曲线的置信带。通过三种方法算例的对比,验证了Bootstrap方法在小子样空间下计算疲劳寿命置信区间的优势。
The Bootstrap method is a way which renewable samples is used to estimate the distribution of unknown statistic. By using the sampling techniques of Bootstrap method, the P-S-N curve of different reliability and its corresponding confidence interval can be got. Based on fatigue specimen data of high strength reinforced flash butt , and using the calculation of three methods of T distribution adjustment method, ASTM method and the Bootstrap method, the confidence interval of P-S-N curve is obtained. Through the comparison of three methods, it verified the advantages of Bootstrap method to calculate the confidence interval of fatigue life under small sample space