Application Research of Motor Faults Diagnosis Based on the Multi-layer NN Structure
Jiancheng Liu · Natural Science Journal of Xiangtan University · 2002
The paper uses the motor current noise multi- varieties time series model parameter Φ i as the initial pattern vector of motor faults category recognition through analysis of its mechanism. In the light of trait of the model parameter, puts forward a fault diagnosis model based on the multi-layer NN structure. In the model, using APEX network extracts classification information and condenses vector space dimensions, making use of feedforward network establishes the classification recognition function. The diagnosis way is correct in practicing.