Bearing Fault Diagnosis Based on Small Sample Learning of Maml–Triplet

Since the emergence of artificial intelligence and deep learning methods, fault diagnosis bearings in rotating machinery has gradually been realized, reducing high costs bearing faults. However, actual work equipment, faults rarely occur, resulting less data. Therefore, it is necessary to study small sample For case data, Maml–Triplet classification framework based on combination maml triplet neural network proposed. In Maml-Triplet classification, firstly, an initial signal feature extractor obtained using Maml training method. Secondly, vectors corresponding data are depth distance measurement network, type judged unknown signal. The results show that accuracy model 2% higher than alone 5% Maml–CNN meta When there fewer samples, gap more obvious. excellent identification ability.