融合物理先验与自监督学习的水平井压裂分布式光纤盲去噪方法

Blind Denoising Method for Distributed Optical Fiber Data of Horizontal Well Fracturing Based on Physics-Informed Self-Supervised Learning

摘要

针对水平井水力压裂过程中分布式光纤声波（DAS）和温度（DTS）数据信噪比低, 且缺乏纯净监测数据作为训练标签的难题, 提出了一种融合物理约束与工程先验的自监督盲去噪方法。该方法无需预先获取无噪真值, 而是利用时-深子块的棋盘格交替采样策略构造自监督训练样本, 实现网络对噪声分布的自适应学习。在此基础上, 针对DTS和DAS的信号特性, 分别引入热传导平滑性、边缘保持及波场相干性等物理约束, 并依据射孔簇位置建立空间变权损失函数, 以在去噪的同时重点保护簇周关键响应特征。该方法在信噪比（SNR）与结构相似性（SSIM）指标上均优于传统滤波, 去噪后的数据显著增强了进液位置的冷锋边界与声波能量条带清晰度, 有效提升了压裂事件识别与簇间进液非均匀性评价的准确性。

Abstract

To address the challenges of low signal-to-noise ratio （SNR） and the lack of clean ground truth labels for distributed acoustic sensing （DAS） and distributed temperature sensing （DTS） data during hydraulic fracturing, a physics-informed self-supervised blind denoising method is proposed. Instead of relying on clean reference data, this method constructs self-supervised training samples using a checkerboard alternating sampling strategy on time-depth sub-blocks and enables the denoising network to learn noise distributions adaptively. Furthermore, physical constraints such as thermal conduction smoothness, edge preservation, and wavefield coherence are introduced for DTS and DAS characteristics, respectively. A spatially weighted loss function based on perforation cluster positions is also established to prioritize the preservation of key response features near clusters. It is demonstrated that the proposed method outperforms traditional filtering methods in terms of SNR and structural similarity （SSIM）. The denoised data significantly enhances the clarity of cooling fronts and acoustic energy bands at fluid entry points, which effectively improves the accuracy of fracturing event recognition and cluster efficiency evaluation.

关键词

分布式光纤传感 / 水平井压裂 / 盲去噪 / 自监督学习 / 物理约束 / DAS/DTS

Key words

distributed fiber sensing / horizontal well fracturing / blind denoising / self-supervised learning / physical constraints / DAS/DTS

参考文献

[1] LUO H, LI H, TAN Y, et al. A novel inversion approach for fracture parameters and inflow rates diagnosis in multistage fractured horizontal wells[J]. Journal of Petroleum Science and Engineering, 2020, 184: 106585.

[2] COSCETTA A, CATALANO E, DAMIANO E, et al. A review of strain-distributed optical fiber sensors for geohazard monitoring：an update[J]. Sensors, 2025, 25(20): 6442.

[3] XU L, SHI S, HUANG Y, et al. Quantification and assessment of steel pitted corrosion using optical frequency domain reflectometry （OFDR）-based distributed fiber optic sensors[J]. Measurement, 2025, 256(Part E): 118519.

[4] SHENG L, LI L, HU L, et al. Distributed fiberoptic sensor for simultaneous temperature and strain monitoring based on Brillouin scattering effect in Polyimide-Coated fibers[J]. International Journal of Optics, 2020, 2020(1): 8810986.

[5] FEO G, SHARMA J, KORTUKOV D, et al. Distributed fiber optic sensing for real-time monitoring of gas in riser during offshore drilling[J]. Sensors, 2020, 20(1): 267.

[6] NI J, WANG C, SHANG Y, et al. Distributed fiber-optic acoustic sensing for petroleum geology exploration[C]// Journal of Physics：Conference Series. 2018, 1065(25): 252029.

[7] WU H, QIAN Y, ZHANG W, et al. Feature extraction and identification in distributed optical-fiber vibration sensing system for oil pipeline safety monitoring[J]. Photonic Sensors, 2017, 7(4): 305-310.

[8] JIANG H, LI J, MA T, et al. Events recognition method in distributed fiber bragg grating system based on residual mixed convolutional neural network[J]. Optics Communications, 2025, 596: 132518.

[9] LI Y, PAN Y, KHODAEI S Z. Real-time deformation digital twin of composite plates via distributed fiber optic sensing [J]. Advanced Engineering Informatics, 2026, 69(Part C): 103992.

[10] ZHOU Y, WU J, CHEN J F, et al. Intermittent monitoring of long-term concrete creep and shrinkage using OFDR-based distributed optical fiber sensors[J]. Measurement, 2025, 258(Part C): 119348.

[11] WEI Z, LIN Z, YAN Q, et al. An improved phase unwrapping method based on deep learning for optical fiber sensors[J]. Optics Communications, 2026, 598: 132648.

[12] HOU G, ZHANG S, SHAO Y, et al. Application of distributed optical fiber technology for strain and temperature monitoring in mass concrete raft foundations[J]. Optical Fiber Technology, 2026, 96: 104491.

[13] FANG Z, WEN Z, FENG Y, et al. Image-assisted 2D partition denoising method for distributed optical fiber vibration measurements[J]. Measurement, 2025, 256(Part E): 118519.

基金资助

中国石油天然气股份有限公司重大科技专项“砾岩油藏规模增储上产与提高采收率关键技术研究”(2023ZZ24)

新疆油田公司重大科技专项“井下特色作业技术研究与应用”(2024XJZD07)