基于粗糙集的转子裂纹故障诊断研究

摘要

针对诊断转子裂纹故障时出现的许多不确定性的信息,信息的重要度不同以及冗余等问题,提出了一种基于粗糙集的裂纹故障诊断方法.利用粗糙集相对约简的不唯一性和不确定知识的表达能力,将诊断转子裂纹的相关特征参数进行模糊化及离散化处理,构建故障决策表.对故障征兆参数进行约简,去除冗余信息,求出集合的正域.根据依赖度和重要度关系,求出集合的核,作为诊断转子裂纹故障的规则,简化了诊断过程.实例表明,该方法能较大的提高转子裂纹故障诊断的效率和可靠性.%In the light of the problems of uncertainty information, important degree different and information redundancy, relating to the Rotor crack fault diagnosis, proposed a crack fault diagnosis method based on rough sets. By utilizing the non-uniqueness of the relative reduction, and the uncertainty knowledge expression ability of the rough sets, the relevant characteristic parameters of the crack rotor were discrete and fuzzed, constructing fault decision table. Then reducing the scale of the fault characteristic parameter, remove redundant information, obtain positive region of set. According to the dependency and importance relationship,obtain positive region Q -core,as the rotor crack fault diagnosis rules,Simplify the diagnosis of process. A practical example has proven that the efficiency and reliability of crack rotor fault diagnosis based on rough sets method.