摘 要: 以自组织特征映射网络SOM和反向传播网络BP为理论基础，提出了SOM-BP神经网络模型，利用UCI数据作为训练样本建立诊断模型。在matlab环境下，通过仿真试验并与单一BP网络进行比较，验证了此复合神经网络用于风机故障诊断的正确性和精确性，并具有易于实现的优点。
关键词：风机 BP神经网络 SOM神经网络 故障诊断
Abstract: The SOM-BP network model based on the principal of self-organization network(SOM) and back pressure network(BP) was established. The faults and relative parameters of UCI for Mechanical analysis were used for the training samples of the
[登陆后可查看全文]
