45 / 2025-03-28 22:13:20

An fault diagnosis method based on self-organizing map neural networks for the artillery automatic machine

Fault diagnosis; Self-Organizing Map; Kullback-Leibler; Automatic machine

Aiming at the challenges in detecting faults of rotating chamber bushings, curved sliding plates, and cam mechanisms in artillery automatic mechanisms, this paper proposes a fault diagnosis method based on Self-Organizing Map (SOM) neural networks. Signals from both faulty and normal states of the rotating chamber driving system are decomposed using Ensemble Empirical Mode Decomposition (EEMD). Fault feature vectors are then extracted through the Kullback-Leibler (K-L) divergence method. These feature vectors are utilized as training samples for SOM-based clustering analysis, enabling effective fault diagnosis of the driving system. Simulation results demonstrate that the proposed algorithm achieves accurate identification of faults in rotating chamber bushings, curved sliding plates, and cam mechanisms within artillery automatic mechanisms.