小波包-神经网络在汽轮机转子故障诊断的应用  

Application of Wavelet Packet-Neural Network to Fault Diagnosis of Turbine Rotors

作　　者：刘明利[1] 傅行军[1] 李艳[1] 

机构地区：[1]东南大学火电机组振动国家工程研究中心,南京210096

出　　处：《发电设备》2009年第6期397-399,共3页Power Equipment

摘　　要：根据汽轮机转子振动信号特点,提出了小波包分析和概率神经网络相结合的故障诊断方法。利用小波包对故障信号进行分解,然后将归一化后的数据用于概率神经网络,从而对信号特征及其故障类型建立非线性映射,实现故障诊断。MATLAB的实验仿真表明,小波包分析和概率神经网络的结合在汽轮机转子常见故障的诊断中是很有效的。According to the signal features of turbine rotor vibration faults,a diagnosis method in combination of wavelet packet and probabilistic neural network is being proposed.Fault signals are decomposed using wavelet packet,and then the unified data will be processed by probabilistic neural network,whereafter a nonlinear mapping relationship between signal features and fault type can be obtained,and thus a fault diagnosis competed.Experimental results of MATLAB simulation show that the combined method by wavele...

关 键 词：汽轮机 转子 故障诊断 小波包 神经网络 

分 类 号：TK268[动力工程及工程热物理—动力机械及工程]