Structural damage acoustic emission signal identification method and device, and storage medium

Abstract

The invention provides a structural damage acoustic emission signal identification method and device, and a storage medium. The method comprises the following steps of: S1, collecting labeled signal samples including structural acoustic emission damage, vibration and friction noise; S2, designing a convolutional neural network model; and S3, carrying out model training and optimization by using a labeled data set. According to the method, an acoustic emission damage signal extraction means based on a machine learning method is adopted, a neural network model is constructed and trained to summarize relatively complex mode characteristics which are difficult to extract manually in the signals, and finally, damage signals are separated from noise in actual analysis according to the mode characteristics. According to the invention, by designing and training the model, the intelligent agent can automatically summarize the generality of samples of the same category and the symbolic difference of samples of different categories from a large amount of data, so that the acoustic emission signal extraction with strong adaptability and good robustness is realized.

Description

technical field [0001] The invention belongs to the technical field of non-destructive testing and structural integrity monitoring, and in particular relates to a structural damage acoustic emission signal identification method, device and storage medium. Background technique [0002] Non-destructive testing is a key technology to ensure the integrity and reliability of internal key components of large-scale engineering equipment and structures during service. With the development of social productivity, more and more of the above-mentioned equipment and structures will be put into service. If their internal structures are damaged and aged after a long period of service, catastrophic accidents will easily occur. In actual engineering, although there are a variety of technical means to carry out non-destructive testing, the need for in-depth data analysis during the testing process still brings a large workload to on-site technicians. [0003] Traditional non-destructive tes...

AI Technical Summary

Problems solved by technology

This method of artificially selecting the characteristics of acoustic emission signals based on expert experience and setting judgment rules based on engineering experience can only roughly understand the signal strength and frequency distribution, and it is difficult to identify complex signal patterns. Separation from vibration, friction and other noise

It is precisely because of this that the technology is currently only applicable to noisy and clean environments such as laboratories in practice, and it is difficult to apply in engineering environments

Images