摘要
将信息融合技术引入到异步电机的故障诊断中。通过小波分析振动传感器和电流传感器的状态信号,提取故障特征参数,把特征参数作为BP神经网络的输入,运用并行的两个BP神经网络对异步电机进行诊断,再用DS证据理论对诊断的结果进行全局融合,实现对异步电机一种或同时多种故障的准确诊断。通过实验表明诊断结果的可信度显著提高,不确定性明显减少,充分显示了该方法是有效的。
The information fusion technology is introduced to the fault diagnosis of asynchronous motor.Through wavelet analysis vibration sensor and the current state of the sensor signal,extracts fault characteristic parameters,as the input of BP neural network,using two parallel BP neural network diagnosis of asynchronous motor,the results by D-S theory of evidence to the diagnosis of the global fusion,implementation of asynchronous motor is a kind of or various faults of an accurate diagnosis.Experiments show that significantly increases the credibility of diagnosis,uncertainty significantly reduced,fully shows the method is effective.
出处
《组合机床与自动化加工技术》
北大核心
2013年第12期86-89,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
广西矿冶与环境科学实验中心开放基金项目(KH2011YB007)
广西研究生教育创新计划资助项目(2011105960812M24)
关键词
异步电机
信息融合
故障诊断
BP神经网络
D-S证据理论
asynchronous motor
information fusion
fault diagnosis
BP neural network
D-S evidence theory
