基于定性定量知识的故障诊断  

Fault Diagnose Based on Integration of Qualitative and Quantitative Knowledge

作　　者：邵晨曦[1] 张俊涛[1] 范金锋[1] 白方周[1] 

机构地区：[1]中国科学技术大学计算机科学与技术系,合肥230027

出　　处：《计算机工程》2006年第6期189-191,274,共4页Computer Engineering

基　　金：国家自然科学基金资助项目(69974038);2004年安徽省科技计划资助项目

摘　　要：提出一种将检测中的定量信息定性化的方法,并应用QSIM的算法约束的概念,结合区间代数的计算规则,设计了定性与定量结合的诊断系统,缩小诊断空间,增大定性仿真在故障诊断中的应用范围。通过不断扩大先验故障模型,使系统具备一定的学习能力,并以压缩制冷系统为例进行了诊断,验证了系统的正确性。A fault diagnosis system based on the combination of quantitative and qualitative information is provided. The conception of QSIM constraint and the representation of interval algebra are used to transform pieces of quantitative information into qualitative one, and a small diagnosis space is obtained. This method may extend the application range of qualitative simulation in fault diagnosis field. Alter added the priori fault model, the system will have the ability of self-learning. An example of diagnosing compression refrigerating system is presented and the validity of diagnosis is proved.

关 键 词：定性仿真 定景知识 系统模型 故障诊断 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]