摘要
针对证据理论应用在旋转机械故障诊断领域,当传感器获得的振动数据存在高冲突时,诊断的精度偏低的问题,提出了将基于向量余弦相似性度量证据冲突的方法与无量纲指标相结合的故障诊断模型。应用于旋转机械故障诊断实验中,诊断结果表明,大大降低了冲突证据对诊断结果的影响,有效地提高了故障诊断的正确率。
Aimed at the problem ol the lower diagnosis accuracy ol that evidence theory was applied inthe field ol rotating machinery fault diagnosis when the data sensors vibration obtained have high conflict,a lusion lault diagnosis model for rotating machinery based on dimensionless index and a method ofmeasure evidence conflict based on vector cosine similarity. It was applied to rotating machinery faultdiagnosis experiment. The diagnosis results show that it greatly reduces the influence ol the conflictevidence on the diagnosis result and ellectively improves the diagnosis accuracy.
出处
《信息技术》
2016年第7期58-61,共4页
Information Technology
基金
国家自然科学基金项目(61174113
61473331
61473094)
茂名市科技计划项目(201311)
关键词
故障诊断
证据理论
高冲突
无量纲指标
旋转机械
lault diagnosis
evidence theory
high conflict
dimensionless index
rotating machinery
