摘要
一个大型的安全关键系统常常由成千上万的部件所组成,在这类系统中检测错误会耗费巨大的时间和代价。讨论如何根据系统的行为和功能,以及部件之间的关系来建立FPG;如何用形式方法描述FPG,以及使用多重错误检测(MFD2)算法来确定错误的源头,利用错误模式和后果分析方法有选择地在系统部件上放置警报器器。为系统来建立一个错误传递图来检测错误,能够极大的减少系统的重担。实验证明在识别错误的来源时使用多重错误检测算法能够识别出产生单一或多重错误的源节点。
A large security critical system is often composed of thousand of components and the diagnosis of faults in such a system is time-consuming and computationally costly. Discusses how to set up FPG according to the behavior and function of the system as well as the relationship between the components, and uses formal methods to describe it. And how to use multiple error detection algorithm to determine the error source, uses error modes and effect analysis method to place the alarm on components selectively. Modelling and analysing fault propagation by graph is helping to reduce such a burden. The experiment shows that using muhiple error detection algorithm in the error sources identification can identify the source node producing single or multiple errors.
出处
《现代计算机》
2011年第19期21-24,29,共5页
Modern Computer
