摘要
In recent years, artificial neural networks(ANNs) have been successfully used in many fields. In order to improve the accuracy of surveillance and diagnosis for large rotating machinery in chemical processes, first, the time domain signal is transformed into featuressuited for the learning of neural network, then, ANN(BP) is used to extract the features of the field vibrational signal. By means of the features extraced, it is easy to monitor the vibration situation of a machinery unit. Checking with practical vibration data has proved this method to be satisfactory.
In recent years, artificial neural networks(ANNs) have been successfully used in many fields. In order to improve the accuracy of surveillance and diagnosis for large rotating machinery in chemical processes, first, the time domain signal is transformed into featuressuited for the learning of neural network, then, ANN(BP) is used to extract the features of the field vibrational signal. By means of the features extraced, it is easy to monitor the vibration situation of a machinery unit. Checking with practical vibration data has proved this method to be satisfactory.
出处
《化工学报》
EI
CAS
CSCD
北大核心
1993年第1期122-126,共5页
CIESC Journal
关键词
神经网络
回转机械
故障
诊断
artificial neural networks, rotor, vibration, technical diagnosis
