TPS自动生成平台上的模糊故障字典法

Fuzzy approach for fault dictionary used in TPS auto-generate platform

目前数字系统的故障诊断和测试性设计技术已经比较成熟,而模拟系统的故障诊断,尽管起步较早,但进展缓慢。这主要由于参数值的连续变化、反馈和非线性等原因,使得模拟系统的故障诊断相对比较复杂。目前已经提出了多种诊断方法,故障字典法即为其中的一种,它通常用于直流电路硬故障的诊断。由于故障因素和容差因素互相交迭,具有一定的模糊性,这给故障字典的应用带来一定的困难。提出了一种基于模糊故障字典的模拟电路故障诊断方法,用模糊故障字典取代传统的故障字典,适用于模拟电路单故障的故障诊断及故障隔离,便于计算机辅助诊断。

At present, the digital system fault diagnosis and testability design technology have come to maturity, and are widely used in the computer and relative fields. Despite starting early in the analog circuit fault diagnosis, it has developed slowly because of some factors, for instance, the parameter valuecontinuously changes, the feedback, non-linearsystem and so on, which resulted in themore complex for analog circuit fault diagnosis. There are a lot of methods suggested to solve such problem. Fault dictionary is usually used for DC diagnosis of analog circuits in single and hard fault conditions. Since every component had a tolerance, the observed response had some deviation from the simulated response. They could not match exactly. This made the diagnosis problem much more difficult. An automatic fault diagnosis technique based on a fuzzy approach is presented. The proposed technique was applied to the detection and the isolation of single soft faults in analog electronic circuits by using a fuzzy fault dictionary instead of usual fault dictionary.