中文摘要：

为了充分利用已有的长期监测得到的元件故障记录数据，根据其数据稀疏且离散的特点，在作者提出的连续型空间故障树（CSFT）基础上，提出适合于处理这些数据的离散型空间故障树（DSFT）．首先对元件故障记录数据分别按照元件的工作时间t和工作温度c进行统计，然后分别在t和c方向投影且归一化数据，最后对故障概率分布点进行拟合得到特征函数，进而得到元件故障概率空间分布．研究表明：DSFT是由表及里地研究元件的故障概率空间分布．将CSFT与DSFT结合分析得到，除t和C外仍有其它环境因素影响着元件故障概率空间分布，并得到了影响范围和特征．

英文摘要：

In order to make full use of existing long-term monitoring of component fault recording data, considering the characteristics of the data sparse and discrete- ness, based on the continuous space fault tree （CSFT） by the author, the discrete space fault tree （DSFT） is put forward for processing the data. Firstly, fault record data are statistical respectively according to the work time t and working temperature c of component, then respectively in the t and c direction projection and normalized data. Finally, the characteristic function is fitted with the points of the failure probabil- ity distribution; the spatial distribution of component failure probability is obtained. DSFT studied on the failure probability spatial distribution of the components is from outside into inner. Combining the CSFT with DSFT, it is analyzed that in addition to the t and c, there are other environmental factors affecting the component failure probability spatial distribution, and then the scope and characteristics are obtained.