CAJ | 학술논문

利用粗糙集能够约简决策表减少得到诊断规则的复杂度和案例分析能够快速计算出两个案例的相似度，本文设计了一种基于粗糙集理论电机故障诊断方法。首先结合粗糙集中区分矩阵的知识消除掉收集的电机故障数据中的冗余信息，以求得到最简决策表，从而得到诊断规则；然后借鉴属性重要度的计算方法，来确定各属性权重，标记出各属性的重要程度，并为计算属性相似度做基础；最后利用案例推理的理论计算测试规则和所得的诊断规则的各相似度，根据最高相似度的诊断规则得出测试规则的故障检测结果。实例验证所得结果也证明了本文设计的方法的可行性与有效性。
이용조조집능구약간결책표감소득도진단규칙적복잡도화안례분석능구쾌속계산출량개안례적상사도，본문설계료일충기우조조집이론전궤고장진단방법。수선결합조조집중구분구진적지식소제도수집적전궤고장수거중적용여신식，이구득도최간결책표，종이득도진단규칙；연후차감속성중요도적계산방법，래학정각속성권중，표기출각속성적중요정도，병위계산속성상사도주기출；최후이용안례추리적이론계산측시규칙화소득적진단규칙적각상사도，근거최고상사도적진단규칙득출측시규칙적고장검측결과。실례험증소득결과야증명료본문설계적방법적가행성여유효성。
Using to reduction of decision table can reduce the complexity and the case analysis the diagnosis rules to quickly calculate the similarity of two cases in rough set theory,this paper designs a kind of motor fault diagnosis method based on rough set theory.First of all,combined with rough set discernibility matrix to eliminate redundant information knowledge in motor fault data collection,in order to get the simplest decision table,thus the diagnosis rules;then from the calculation method of attribute importance degree,to determine attribute weights,mark out theheavy d egree to each attribute,and make a foundation for the attribute similarity;the similarity calculatio n ofdiagnosis rules finally test rules and income using the case-based reasoning theory,according to the fault detectiondiagnosis results obtained the highest similarity rules for test rules.An example to validate the obtained results also proved the feasibility of this design method,and the effectiveness of the.