Aero-engine transmission system fault diagnosis method based on domain adaptive graph convolutional network

Abstract

The invention discloses an aero-engine transmission system fault diagnosis method based on a domain adaptive graph convolutional network. The method comprises the steps of: collecting first vibration signals of an aero-engine transmission system with an unknown fault at different rotating speeds, and taking the first vibration signals as unlabeled target domain data; constructing a domain adaptive graph convolutional network, collecting second vibration signals of an aero-engine transmission system with a known fault at different rotating speeds, taking the second vibration signals as labeled source domain data, and taking the source domain data and part of the target domain data as input to train the domain adaptive graph convolutional network; and inputting the other part of target domain data into the trained domain adaptive graph convolutional network to obtain a prediction label of the target domain data. Migration diagnosis of a fault of an aero-engine transmission system at different rotating speeds is realized.

Description

technical field [0001] The disclosure belongs to the field of mechanical fault diagnosis, and in particular relates to a fault diagnosis method for an aircraft transmission system based on a domain-adaptive graph convolution network. Background technique [0002] Intelligent diagnosis plays an important role in the health management system of aero-engine transmission system, which has been widely used in modern industry, and its main purpose is to monitor the status of equipment and reduce downtime. At present, unsupervised domain adaptive methods have been successfully applied in mechanical fault diagnosis under variable working conditions. In unsupervised domain adaptation methods, three types of information, such as class label, domain label and data structure, are crucial to achieve the process of transferring from a labeled source domain to an unlabeled target domain. However, most existing unsupervised domain adaptation methods only use the first two kinds of informat...

AI Technical Summary

Problems solved by technology

However, most of the existing unsupervised domain adaptation methods only use the first two kinds of information and ignore the modeling of the data structure, which makes the information contained in the features extracted by the deep network incomplete.

Images