Wind generating set gearbox diagnosis method based on semi-supervised momentum prototype network

• Main Authors: LUO MAOLIN, YU HONG, SU ZUQIANG, ZHANG XIAOLONG, YU JIANHANG, HAN YAN
• Format: Patent
• Language: Chinese; English
• Published: 11.01.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention relates to a wind generating set gearbox diagnosis method based on a semi-supervised momentum prototype network, and belongs to the field of fault diagnosis. The method comprises two processes of semi-supervised momentum prototype network training and fault diagnosis and identification. The method comprises the following steps: 1, training a semi-supervised momentum prototype network; and 2, carrying out diagnosing the fault of the gearbox of the wind generating set. The invention provides a fault diagnosis method for the gearbox of the wind generating set based on the semi-supervised momentum prototype network by combining pseudo-label learning in semi-supervision and the semi-supervised momentum prototype network in small sample learning. According to the method, the fault information contained in the unmarked sample is fully utilized, the over-fitting phenomenon of the prototype network caused by scarcity of the marked sample is greatly reduced, and the stability and accuracy of the semi-supe