一种基于多级信息融合技术的系统故障诊断方法  

An Approach to Fault Diagnosis Based on Hierarchical Information Fusion

作　　者：刘宜平[1] 沈毅[1] 童树鸿[1] 刘志言[1] 

机构地区：[1]哈尔滨工业大学控制工程系,150001

出　　处：《系统工程与电子技术》2000年第12期97-100,共4页Systems Engineering and Electronics

基　　金：国家自然科学基金!资助课题 (6 990 40 0 4)

摘　　要：采用多级信息融合思想 ,利用系统故障症状的分散性 ,提出了一种新的系统故障诊断方法。首先 ,我们采用多传感器同源数据融合来保证量测信号的准确可靠 ;然后 ,利用同一故障在系统中的不同表现症状 ,采用模糊积分融合 ,融合输出为相对应的故障诊断结果。最后对燃气轮机的多部位故障诊断进行了仿真研究 ,结果较为理想。In this paper, we propose a novel approach to fault diagnosis based on hierarchical information fusion. First, we use the same source data fusion to confirm the reliability and accuracy of the measured signals. Then, we adopt multiple neural networks as fault classifiers to obtain the local decisions. Furthermore, we present a method for decision fusion based on fuzzy integral in which the relative importance of the different individual network is considered. Finally, we apply the approach to a turbine system and the experimental results show its good performance.

关 键 词：故障诊断 系统可靠性 神经网络 信息融合 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]