Removing multiple types of noise of distributed acoustic sensing seismic data using attention-guided denoising convolutional neural network

Abstract

In recent years, distributed optical fiber acoustic sensing (DAS) technology has been increasingly used for vertical seismic profile (VSP) exploration. Even though this technology has the advantages of high spatial resolution, strong resistance to high temperature and pressure variations, long sensing distance, DAS seismic noise has expanded from random noise to optical abnormal noise, fading noise and horizontal noise, etc. This seriously affects the quality of the seismic data and brings huge challenges to subsequent imaging, inversion and interpretation. Moreover, the noise is more complex and more difficult to simultaneously suppress using traditional methods. Therefore, for the purpose of effectively improving the signal-to-noise ratio (SNR) of DAS seismic data, we introduce a denoising network named attention-guided denoising convolutional neural network (ADNet). The network is composed of four blocks, including a sparse block (SB), a feature enhancement block (FEB), an attention block (AB) and a reconstruction block (RB). The network uses different kinds of convolutions alternately to enlarge the receptive field size and extract global feature of the input. Meanwhile, the attention mechanism is introduced to extract the hidden noise information in the complex background. The network predicts the noise, and denoised data are obtained by subtracting the predicted results from the noisy inputs. In addition, we uniquely construct a large number of complex forward models for pure seismic data training set to enhance the network suitability. The combination design improves the denoising performance and reduces computational cost and memory consumption. The results obtained from both synthetic- and field data illustrate that the network has the ability to denoise the seismic images and retrieve weak effective signals better than conventional methods and common networks.

Multiple Noise Reduction for Distributed Acoustic Sensing Data Processing through Densely Connected Residual Convolutional Networks

“…In this study, considering the high accuracy and resolution required for the geological model, the synthetic dataset is chosen to generate DAS data using Tesseral software (Wang et al, 2023), which can handle multi-scale models. The horizontal distance is set to 600 m, and the vertical depth to 1000 m. We construct synthetic DAS data with 600 traces and a 2 ms sampling rate for testing, as shown in Fig.…”

“…To demonstrate the denoising performance of our proposed method on synthetic and real-time data, we chose the bandpass filtering method (Wang et al, 2023) and two classical deep learning architectures as benchmark methods:…”

“…In addition, DAS generally exhibits lower microseismic and seismic detectability compared to conventional geophones (Karrenbach et al, 2019). DAS can utilize a variety of light sources to enhance performance, such as rectangular pulses (Pastor-Graells et al, 2016), chirped pulses (Fernández-Ruiz et al, 2017), and Optical Frequency Combs (Zhang et al, 2023;Huang et al, 2024;Huang et al, 2023). The signal-to-noise ratio (SNR) of genuine DAS data signals is much lower than the counterpart, thus making improving the SNR of DAS data an essential task.…”