Cross-condition Fault Diagnosis Algorithm for Rocker Arm Transmission System Based on Parameter Transfer
The rocker arm transmission system plays a crucial role in shearer,which directly affects the efficiency of coal production.Aiming at the difficult problem of cross-condition transfer fault diagnosis,research has found that the fault data feature distributions between the source domain and target domain are similar.Therefore,a fault diagnosis model based on parameter transfer was proposed.This model adopts the transfer convolutionl neural networks with global average pooling(TCNNG)model,which automatically extracts feature information from source domain condition data.By a parameter transfer and fine-tuning strategy,the diagnostic knowledge parameters are successfully transferred from the source domain to the target domain,which significantly improves the fault recognition rate and generalization performance of the model under different conditions.
shearerrocker arm transmission systemfault diagnosisdeep learningtransfer learning
NETL
NSTL
万方数据
