基于神经网络和D-S证据理论的信息融合故障诊断方法

Fault Diagnosis Method of Data Fusion based on Neural Network and D-S Evidence Theory

为了解决齿轮传动系统检测难度大、准确性不高和多点测试时信息处理复杂的问题,提取振动信号统计量特征参数、利用神经网络技术与D-S(Dempster-Shafert)证据理论相结合的信息融合故障诊断方法,实现了数据级、特征级与决策级的多级融合诊断。实验结果表明,将信息融合方法用于齿轮传动系统故障诊断,有助于综合利用故障信息,提高了故障诊断的准确性和可信度。

Aiming to resolve the problems of test difficulty,low accuracy and complex information processing lay in gearbox fault diagnosis,vibration signal feature parameter is extracted,neural network and D-S evidence theory are combined to accomplish multi level data fusion diagnosis.Experiment result indicates that the data fusion method using in gear transmission system fault diagnosis can improve the accuracy and reliability of fault diagnosis.