Rotor imbalance fault diagnosis method based on convolutional neural network

• Main Authors: LI WEIJUN, ZHANG KAI, YING GUANGYAO, LIU RUONAN, CAI WENFANG, CHEN DONGYUE, WU WENJIAN, ZHANG BAO
• Format: Patent
• Language: Chinese; English
• Published: 11.01.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a rotor imbalance fault diagnosis method based on a convolutional neural network, and the method comprises the following steps: step 1, collecting axis trajectory data, comparing the axis trajectory data with standard axis trajectory data, and analyzing the fault type of the axis trajectory data; step 2, establishing a convolutional neural network model, wherein the convolutional neural network model sequentially comprises three convolutional layers, a pooling layer, a full connection layer and a Softmax layer; and step 3, dividing the axis trajectory data into a training set and a test set in proportion, training the convolutional neural network model on the training set, testing by using the test set, performing multiple rounds of training, and searching an optimal model hyper-parameter. Based on the morphological features of the axis trajectory, the features of the rotor trajectory image data are automatically extracted through the strong feature extraction capability of the convolu