Gearbox fault diagnosis method, system and device based on double attention mechanism

The invention discloses a gearbox fault diagnosis method, system and device based on a double attention mechanism, and the method comprises the following steps: S1, obtaining an original vibration signal of a gearbox, and obtaining a multi-dimensional data set; s2, performing time-frequency conversion on the original vibration signal of the gearbox to obtain two-dimensional image data; s3, inputting the two-dimensional image data into a DAnet model, respectively gathering self-attention in a space attention module and a channel attention module, and then performing fusion to obtain a final feature map; and S4, carrying out fault identification on the final feature map through a full connection layer and a softmax classifier. According to the invention, the fault type classification of the gearbox can be realized, the fault characteristic effect is obviously enhanced, the method has certain anti-interference capability, and the accuracy of fault identification can be improved. 本发明公开了基于双注意力机制的齿轮箱故障诊断的方法、系统及装置，其