摘要
研究了螺旋桨叶片疲劳试验中定时截尾数据的处理和评估方法。采用线性Goodman修正和Miner损伤累积法则将截尾疲劳寿命转换为疲劳极限样本数据,对该样本建立似然方程,利用似然方程组的极值条件迭代求解疲劳极限的最大似然估计量,并通过Monte-Carlo假设验证了方法的准确性。根据疲劳极限的最大似然估计可以得到更准确的S-N曲线,从而更有效地进行疲劳寿命评估。
The processing and estimation method of fixed time censoring data in a propeller fatigue test was investigated.The fatigue limit was derived from fatigue test results by linear Goodman modification and Miner damage accumulation rule.The likelihood equations of this sample were established and solved using E-M iteration algorithm,so as to obtain the maximum likelihood estimation. The algorithm was verified employing Monte Carlo inspection. With the maximum likelihood estimation of the fatigue limit,a more accurate S-N curve was obtained for life assessment of the blade.
出处
《现代机械》
2016年第5期31-34,共4页
Modern Machinery
