论文部分内容阅读
针对航空机载设备可靠性增长试验数据不规则的特点,在AMSAA模型的基础上提出了识别异常点的AMSAA模型与处理截断数据的AMSAA模型,并给出了模型拟合优度检验方法。经实例验证表明,识别异常点的AMSAA模型可以在给定置信度下识别出异常点并排除异常点对瞬时MTBF(平均故障间隔时间)极大似然估计值的干扰;截断数据的AMSAA模型能够利用被截取的部分数据,得到准确的瞬时极大似然估计值。
AMSAA model for identifying abnormal points and AMSAA model for handling truncated data are proposed based on the AMSAA model. A test method for the goodness of fit of the model is also given. The experimental results show that the AMSAA model can identify the anomaly under the given confidence level and eliminate the interference of the anomalous point to the maximum likelihood estimation of the instantaneous MTBF (MTBF). The AMSAA model of the truncated data can Using the partial data that is intercepted, an accurate instantaneous maximum likelihood estimation is obtained.