论文部分内容阅读
用遗传算法建立了汽机故障诊断的人工神经网络模型，以小波包分解技术获得的１０个频段上的能量为网络的输入模式，对汽机常见的几种故障进行分类训练，并应用于待识别故障样本的识别计算，结果表明该方法在汽机故障诊断中是有效的。
The artificial neural network model of steam turbine fault diagnosis is established by using genetic algorithm. The energy of the 10 frequency bands obtained by wavelet packet decomposition is taken as the input mode of the network. Several common faults of the turbine are classified and trained, and applied to the fault to be identified The results show that this method is effective in turbine fault diagnosis.