Train axle fault acoustic emission detection method based on TCN network

Abstract

The invention discloses a train axle fault acoustic emission detection method based on a TCN network. The method comprises the following steps: constructing the TCN network; inputting a large number of samples into a TCN network for network training, wherein the samples comprise acoustic emission signals, wheel shaft rotating speeds and corresponding classification labels; the TCN performs feature extraction on the acoustic emission signals, considers the correlation between the acoustic emission signals and the wheel shaft rotating speed, combines the learned features with the wheel shaft rotating speed to further perform feature learning, outputs a classification result, continuously performs network training, and stops training when a preset training frequency is reached or a loss function value of the network is minimum; a trained TCN network model is obtained; inputting the acoustic emission signal detected in real time and the corresponding axle rotating speed into the trained TCN network model to obtain a train axle fault detection result; the method has the advantages that data information is fully utilized, and information loss caused by manual feature selection is avoided.

Description

technical field [0001] The invention relates to the field of train wheel axle fault detection, and more particularly to a train wheel axle fault acoustic emission detection method based on a TCN network. Background technique [0002] The long-term operation of the train, the long-term mechanical rotation of the train axle and the huge load, is very easy to cause wear and tear or even failure, which may cause major accidents and cause major economic losses. Therefore, the wear and failure of the axles can be found in time to maintain and replace them, which can effectively avoid train accidents. It is of great significance to detect the fault of the train axle in time. [0003] At present, the fault diagnosis methods of train axles are mainly divided into two categories. One is a contact fault diagnosis method based on vibration signals, and the other is a non-contact diagnosis method based on acoustic features. The wheel and axle fault diagnosis based on vibration signal ...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is that the existing acoustic emission detection method for train axle faults mainly relies on manual methods to find fault characteristics, which is likely to cause information loss

Images