航空学报 > 2000, Vol. 21 Issue (4): 343-345

智能故障模式影响分析的推理研究

赵廷弟, 杨为民   

  1. 北京航空航天大学可靠性工程研究所,北京100083
  • 收稿日期:1999-04-09 修回日期:1999-09-01 出版日期:2000-08-25 发布日期:2000-08-25

STUDY ON REASONING FOR INTELLIGENT FAILURE MODE AND EFFECT ANALYSIS

ZHAO Tingdi, YANG Weimin   

  1. Institute of Reliability Engineering, Beijing Univ. of Aero. and Astro., Beijing 100083, China
  • Received:1999-04-09 Revised:1999-09-01 Online:2000-08-25 Published:2000-08-25

摘要:

介绍了应用神经网络和专家系统技术,依托专家知识,基于功能/硬件而建立的功能关系模型图进行推理的智能故障模式影响分析系统。详细论述了其故障模式选择推理模型和应用神经元网络及其扩展技术建立的产品故障模式影响分析推理模型,并对推理流程进行了详细的描述。此外,对其系统结构、各组成部分的相互关系,及系统的功能流程也作了简要的论述。

关键词: 故障模式影响分析, 可靠性, 人工智能, 专家系统, 神经元网络

Abstract:

An intelligent system for Failure Mode and Effect Analysis (FMEA) where the neural network and expert system are applied was introduced in this paper. At first, the structure and relationship of each element and the function procedure were discussed. And then, the FMEA reasoning models were described in detail. The products failure model selecting reasoning model was established by applying the rule based production system model. Using the neural network and its extended technique, the FMEA reasoning model was established. In the model, the activating rule of the nerve cell was a set of many output functions, and this network was a hierarchical and nested network in which each product is a sub network. It can not only express the product hierarchy but also describe the failure model effect relationship.

Key words: failure mode, reliability, artificial intelligence, expert system, neural network

中图分类号: