The Fault Diagnosis Of The Blower Fan Based On The Neural Network

With the development of equipment complexity and the improvement of automatic level, the subject of equipment fault diagnosis is more important. And that, the artificial neural network that imitate physical structure of cerebella and be strong able to operate together and associate is propitious to diagnose equipment fault. Contraposed the system of fault diagnosis and inspection about blower fan discussed the related theory to the equipment vibration fault diagnosis and made use of a synthetic artificial neural network, the papers discuss the fault diagnosis of the blower fan. At first, in the paper, the structure of the back propagation neural networks is presented and improves on it, which is well to diagnose. And then, the fault characteristic of the rotated machinery and the fault mechanism of the blower fan are analyzed, to summarize the symptom table about the blower fan. Two vibration sub networks are constructed, and the faults are analyzed about the blower fan by integration neural networks, which are realized with the Matlab programs. In the last, according to the examples and analyses , this paper confirm that the integration BPNN can recognize kinds of familiar faults of blower fan truly and quickly, and it will enhance the veracious degree and reduce to the human's dependence degree if it is adopted that the neural networks diagnose the fault.