矿业科学学报2026，Vol.11Issue(3)：499-509,11.DOI:10.19606/j.cnki.jmst.2025107
基于WT-U-Net网络与地震图像的断层识别方法
Fault identification method based on WT-U-Net network and seismic images
摘要
Abstract
Fault prediction is critical for safe coal mine production.Traditional machine learning meth-ods suffer from poor prediction accuracy when fault features are subtle.This study therefore proposes a WT-U-Net model by combining wavelet transform(WT)with U-Net to improve interpretation accuracy.Fault-related attributes were extracted from post-stack seismic data,where four attributes with low mu-tual dependence were identified through correlation analysis.The decomposition-reconstruction errors and energy differences of different wavelet basis functions applied to seismic data were then compared.The coif3 mother wavelet was selected for fault detection as its wavelet transform amplified fault-related signatures.The U-Net model was constructed to predict faults in the study area.Results demonstrate that the WT-U-Net model showed higher prediction accuracy than UNet alone on real datasets,with outputs more consistent with manual interpretations and improved convergence.The model also exhibi-ted robustness and generalization in blind tests across other regions.The application of wavelet transform in seismic data denoising enhances fault-related signals,thereby increasing the model's accuracy.This study offers a new solution for intelligent fault identification in coal mining applications.关键词
U-Net/解释/小波变换/机器学习/地震属性Key words
U-Net/explanation/wavelet transform/machine learning/seismic attributes分类
矿业与冶金引用本文复制引用
孙培榛,邹冠贵,朱国维,曹益玮,荣志鸿..基于WT-U-Net网络与地震图像的断层识别方法[J].矿业科学学报,2026,11(3):499-509,11.基金项目
国家重点研发计划(2023YFB3211002) （2023YFB3211002）
国家自然科学基金(42274165) （42274165）
