Helicopter fault diagnosis method, system and equipment based on graph convolutional network

• Main Authors: WU XUDONG, YU TIANYU, XIE ZHIJIE, LI XUEYI, LI DAIYOU
• Format: Patent
• Language: Chinese; English
• Published: 05.07.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a helicopter fault diagnosis method, system and equipment based on a graph convolutional network, and belongs to the application of rotating machinery fault diagnosis in industrial practice, and the system comprises six steps of data acquisition, data processing, data division, method construction, model training and model testing. In the data acquisition stage, sensors are used for acquiring vibration signals of parts such as a transmission system, a power device, a rotor system and a tail rotor system of the helicopter, in the data processing stage, interpolation processing and normalization operation are carried out, and in the data division stage, processed data is divided into a training set and a test set; according to the method, a Gaussian filter, a denoising self-encoding module and a multi-receptive field graph convolutional network are constructed and fused, and a training set and a test set are used for training and testing in a model training stage and a model testing stag