Rolling bearing fault diagnosis method based on WOA-VMD and GAT

• Main Authors: YANG HUIMIN, ZHANG SHENG, FAN YUQI, WANG YAPING, ZHANG QISONG, CAO RUOFAN
• Format: Patent
• Language: Chinese; English
• Published: 29.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a rolling bearing fault diagnosis method based on WOA-VMD and GAT, and the method comprises the following steps: 1, carrying out the self-adaptive determination of the number k of parameter modes decomposed by VMD and a penalty parameter alpha through a WOA optimization algorithm, carrying out the VMD decomposition of an original signal, and carrying out the self-adaptive determination of the number k of parameter modes decomposed by VMD; then, adopting Pearson correlation analysis on the decomposed signals to screen out IMF components with large correlation so as to reconstruct the signals and complete signal noise reduction; and step 2, Attention and graph convolution operation are combined, a graph attention neural network rolling bearing fault diagnosis model is constructed, and more proportions are allocated to value information so as to optimize an information collection stage of a constructed graph and improve the accuracy of model fault diagnosis. Experimental verification show