多级混合数据融合在回转机械诊断中的应用  

Application of hierarchical information fusion to rotating machinery diagnostics

作　　者：杨春燕[1] 朱茹敏[1] 李长诗[1] 

机构地区：[1]郑州轻工业学院,机电工程学院河南郑州450002

出　　处：《郑州轻工业学院学报（自然科学版）》2005年第3期84-87,共4页Journal of Zhengzhou University of Light Industry:Natural Science

摘　　要：根据大型回转机械现场实际,采用数据层、特征层和决策层3层次的混合信息融合技术,建立了空间转子诊断模型,并利用该诊断模型对某大型压缩机组进行了故障诊断.结果表明:采用该诊断模型诊断出的故障与现场发生的故障相吻合,诊断结果具有较高的可信度.Based on the real field engineering of large-scale rotary machine,the paper sets up a spatial rotor＇s diagnosis model through a three hierarchical information fusion of data level and character level and decision level, a compressor unit using the diagnosis model is diagnosed. The result show that the malfunction diagnosed using the model accords with the real field engineering, and the diagnosis result has the higher credence.

关 键 词：信息融合 故障诊断 全矢谱 概率神经网络 

分 类 号：TH165.3[机械工程—机械制造及自动化]