ADCDDA model-based 3D printer fault diagnosis method
The invention discloses a 3D printer fault diagnosis method based on an ADCDA model. The method comprises the steps that source domain 3D printer vibration data are collected, and label classification is carried out; processing the multi-dimensional time sequence data into a multi-channel GASF image...
Saved in:
Main Authors | , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
23.07.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a 3D printer fault diagnosis method based on an ADCDA model. The method comprises the steps that source domain 3D printer vibration data are collected, and label classification is carried out; processing the multi-dimensional time sequence data into a multi-channel GASF image through a sliding window by adopting a Grubm angle field; training a source domain feature extractor and a class discriminator under the condition of domain constraint by adopting the labeled source domain data and the unlabeled target domain data; training a target domain feature extractor and a domain discriminator by adopting the label-free source domain data and the label-free target domain data; and real-time vibration data of a target domain 3D printer are collected, a GASF image is obtained through a Grubm angle field, and the GASF image is input into the trained ADCDA model to obtain a fault diagnosis result. According to the method provided by the invention, the fault diagnosis requirement of the 3D print |
---|---|
Bibliography: | Application Number: CN202410479338 |