摘要
多层前馈人工神经网络在装备故障诊断中的应用含设备运行状态特征值设定和故障判定。并建立神经网络样本、确定网络结构、训练网络等。同时,将状态特征值作为输入样本,得出网络的输出,作为故障类型特征值,最后选定故障决策阈值,根据神经网络输出来判定装备是否出现某种故障。
The application of multi-layer feed-forward artificial neural network in fault equipment diagnosis includes feature value setting of equipment operation condition and fault judgment. The neural network samples were established, the network structure was ensured and the network was trained. At the same time, the condition feature value was taken as the output sample to achieve network output and diagnosis type feature value. At last, the diagnosis decision threshold value was chosen to judge whether the equipment owns diagnosis or not according to the neural network output.
出处
《兵工自动化》
2006年第5期40-41,共2页
Ordnance Industry Automation
关键词
神经网络
状态检修
故障诊断
Neural network
Condition-based maintenance
Fault diagnosis
