线性独立增量过程分析方法

提出了线性独立增量随机过程的概念和定义,建立了线性独立增量过程回归模型及其分析方法,给出了回归方程和高置信度、高可靠度的置信上、下限曲线.传统回归分析方法只能对数据点进行分析处理,线性独立增量过程分析方法则能够对过程进行回归分析,因此可以对性能退化、裂纹扩展等线性独立增量过程进行整体推断,具有信息量大、精度高的特点.大量Monte Carlo模拟和工程应用表明,在试样数相同的情况下,该方法比传统方法具有更高的精度,而在精度相同的情况下,则可减少试样. 

Analysis method for linear process with independent increments

The concept and definition of linear process with independent increments were presented. Then the regression model and analysis method for linear process with independent increments were established. Furthermore, the regression equation and confident limit curves with high confidence and high reliability were given. The proposed method extends the traditional regression analysis, which is only suitable to data points to stochastic processes. Therefore the linear processes with independent increments such as performance degradation curves, crack propagation curves and some other processes, can be analyzed as an integration in this method. Compared with traditional method, this presented method not only has higher precision but also solves the problem of reliability assessment with very small sample. The method has been applied in reliability tests successfully and it shows that a great number of specimens can be saved.