An improved autoencoder for denoising acoustic emission signals in rock fracturing

ABSTRACT

Rock fracture acoustic emission (AE) signals are commonly used non-destructive testing data in geological exploration, resource exploitation, and engineering fields. However, these signals are often accompanied by noise interference caused by environmental factors. In this study, we propose an enhanced model for denoising rock fracture AE signals, called simplified fully convolutional denoising autoencoder (SFCDAE). This model is based on the denoising autoencoder principle in the field of deep learning neural networks. The SFCDAE model consists of only seven layers, with minimal preprocessing of data input. By comparing denoising performance evaluation indicators, higher peak signal-to-noise ratio (PSNR) and lower root mean square error (RMSE) were achieved. On average, PSNR increased by 5.575% and RMSE decreased by 22.225%. Using simulated environmental noise to validate the model, it was found that the model has good robustness and can remove artefacts from sudden noise. The practical application value of the LSTM classification model was validated using data containing real experimental noise, resulting in a higher classification accuracy of 80.083%. These results indicate that the proposed model has better denoising performance compared to existing intelligent models and has certain practical value.

KEYWORDS:

• Acoustic emission
• denoising
• rock fracture
• autoencoder
• deep learning

Disclosure statement

No potential conflict of interest was reported by the author(s).

Announcement

1. We declare that we have no financial and personal relationships with other people or organisations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and company that could be construed as influencing the position presented in, or the review of, the manuscript entitled.

2. We guarantee that the research content is sourced from real and reliable data and information, and that all content has been fully authorised and recognised. The raw data is sourced from existing research (data resource: 10.1029/2021jb022532).

3. We respect intellectual property rights and will never plagiarise, plagiarise, or plagiarise papers, works, etc. All cited content will indicate its source and comply with relevant provisions of copyright law.

Author statement

Author 1 (First Author): Conceptualisation, Methodology, Software, Survey, Analysis, Writing – First Draft;

Author 2: Data organisation, software, writing – first draft;

Author 3 (Corresponding Author): Resources, Supervision;

Author 4: Visualisation, Investigation;

Author 5: Review and editing;

Author 6: Review and editing.

Additional information

Funding

The financial support for this manuscript comes from the National Natural Science Foundation of China [Grant No. 52074088, Grant No. 52174022], Development of western oil fields special project [Grant No.XBYTKT202001], Heilongjiang Province postdoctoral Research initiation project [LBH-Q21086], Basic Research Expenses of Heilongjiang provincial Colleges and Universities: Northeast Petroleum University Control Science and Engineering Team Special Project [2022TSTD-04], ‘Open bidding for selecting the best candidates’ Heilongjiang Province Science and Technology Research Project [DQYT-2022-JS-758], Open Project of National Engineering Laboratory for Exploration and Development of Low Permeability Oil and Gas Fields [KFKT2023-10].