基于混合推理机制的故障诊断专家系统  

Fault Diagnosis Expert System Based on Mixed Reasoning Mechanism

作　　者：何波[1] 刘全利[1] 王越[1] 王华秋[1] 

机构地区：[1]重庆工学院计算机科学与工程学院

出　　处：《微计算机信息》2006年第09S期192-194,共3页Control & Automation

基　　金：重庆市科技重大项目(CSTC.2004AA2001);湖南省机械设备健康维护重点实验室开放基金项目资助。

摘　　要：多数故障诊断专家系统采用单一的推理机制,或者基于规则的推理,或者基于事例的推理。而这两种推理机制都各有优缺点,采用单一推理机制会造成诊断的不准确性。论文将基于规则的推理和基于事例的推理相结合,设计了混合推理机制。在此基础上,论文设计了一个既有专家知识库,又有故障事例库,具有自学习能力的故障诊断专家系统(AFDES)。实验结果表明,论文设计的混合推理机制是比较有效的。Most fault diagnosis expert system adopted single reasoning mechanism, or rule-based reasoning or case-based reasoning. However, each reasoning mechanism had advantages and disadvantages. The adoption of anyone would lead to inaccurate diagnosis. The paper designed mixed reasoning mechanism based on rule-based reasoning and case-based reasoning. On the base of this, the paper designed a self-learning fault diagnosis expert system, namely, AFDES, which had expert knowledge database and fault case database. The experiments indicate that designed mixed reasoning mechanism is effective.

关 键 词：故障诊断 专家系统 基于规则的推理 基于事例的推理 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]