基于规则和本体的汽轮机故障分析  

Steam turbine fault analysis based on rule and ontology

作　　者：郭庆琳[1,2] 张铭[2] 

机构地区：[1]华北电力大学计算机科学与技术学院,北京102206 [2]北京大学计算机系,北京100871

出　　处：《电力自动化设备》2009年第4期11-14,共4页Electric Power Automation Equipment

基　　金：中国博士后科学基金项目(20080430250);国家自然科学基金项目(70572090)~~

摘　　要：将谓词逻辑引入汽轮机故障分析系统中,首先将事实和规则化为与或形,接着采用最一般合一算法对子句集替换合一,最后采用归结原理对汽轮机的故障进行分析。同时,提出采用基于知识本体的推理方法来解决模糊知识的表达和推理问题。采用SPARQL语言和Jena推理机进行模糊知识的推理。SPARQL语言使用SQL语句的查询形式,并提供知识本体查询功能,在本体中,所有的知识都是三元组的形式(〈subject,property,object〉),SPARQL只查询本体库中存储的三元组,没有推理功能。因此,把SPARQL查询嵌入在Jena推理机中,通过推理得到隐含的模糊知识三元组。实现了基于规则和本体的汽轮机故障分析仿真系统,其诊断正确率达到89%。该系统具有很高的效率及容错能力。An approach of steam turbine fault analysis based on rule and ontology is presented. The predicative logic is introduced into the fault analysis system, in which, the facts and rules are formatted with ＂and -or＂ forms,the clause set is unified by the most general unifier algorithm,and the steam turbine faults are analyzed by the inductive reasoning. The ontology-based illation method is used to express the fuzzy knowledge with SPARQL language in the searching form of SQL sentences and to deduce the fuzzy knowledge with Jena Reasoner. In ontology, all knowledge is expressed in triplet form （（subject, property, object））. Without reasoning,SPARQL only searches the triplet saved in the ontology library. The implied triplet is obtained through reasoning by embedding SPARQL enquiry into Jena Reasoner. A steam turbine fault analysis system based on rule and ontology is implemented, which has higher efficiency and error-tolerance capability, and its accuracy is 89%.

关 键 词：汽轮机 故障分析 规则 本体 谓词逻辑 最一般合一算法 归结原理 三元组 

分 类 号：TM311[电气工程—电机] TP206.3[自动化与计算机技术—检测技术与自动化装置]