中文摘要：

为解决DSFT通过因素投影拟合法得到的特征函数无法表示原始离散故障数据分布特征的缺点,引入模糊结构元理论将特征函数结构元化,形成模糊结构元特征函数来解决该问题。模糊结构元特征函数可通过模糊结构元E表示原始数据的离散分布特征,并在计算过程中将E传递至最终结果。通过处理结果中E来得到分析结果的置信度。模糊结构元特征函数可将DSFT问题转化为较成熟的CSFT进行处理,而不用为实现相同功能针对DSFT研究新的方法。论文给出了模糊结构元特征函数的构建思想、基础理论、具体步骤和方法。并针对已有数据和特征函数构建了相应的模糊结构元特征函数。

英文摘要：

In order to solve the faults that characteristic function obtained by factors projection fitting method can not represent the original discrete fault data distribution, introduced the fuzzy structured element theory to transform the characteristic function, form the fuzzy structured element characteristic function to solve the problem. Fuzzy structured element characteristic function by fuzzy structured element E to express the discrete distribution characteristics of the original data, and in the calculation process, the distribution characteristics will be transferred to the final result. Confidence is obtained by analysis E in the results. Fuzzy structured element characteristic function can be DSFT problem into a more mature CSFT processing, without regarding new method of DSFT for the same function of CSFT. This paper presents the construction of ideology, the basic theory, the concrete steps of fuzzy structured element characteristic function, and the construction of the function corresponding for the existing data and the characteristic function.