煤田地质与勘探2026，Vol.54Issue(4)：216-226,11.DOI:10.12363/issn.1001-1986.25.07.0487
多特征联合决策的矿井数据融合技术
Mine data fusion technology based on multi-feature joint decision making
摘要
Abstract
[Background]In the field of geological exploration of coal mines,methods such as in-seam seismic explora-tion,the transient electromagnetic method,and audio-frequency electrical penetration offer unique technical advantages.However,each of these methods can only reflect a single physical property of a medium,suffering from limitations in-cluding the one-sidedness of information and the multiple solutions of interpretations,thus failing to adapt to the new situations of current coal mine exploration.Multi-source data fusion promotes a technological development by leaps from one-sided detection to comprehensive analysis through mechanisms of information complementation,feature en-hancement,and field-source integration.Therefore,this technology can enhance the accuracy of both the positioning and morphological identification of anomalous geobodies.[Methods]The data fusion technology based on wavelet decom-position,which employs principal component analysis(PCA)for low-frequency components and multi-feature joint de-cision-making for high-frequency components,was used to integrate geophysical data obtained using the three methods(i.e.,in-seam seismic exploration,transient electromagnetic method,and audio-frequency electrical penetration).The purpose is to enhance the comprehensive analytical capacity for geological information and to achieve accurate detec-tion of geological structures and anomalies.First,raw data obtained using the three methods were preprocessed to elim-inate noise interference and normalize data formats.Subsequently,using the wavelet transform,the resulting data from various sources were then decomposed into the coefficients of subbands corresponding to different frequencies.Based on their importance in geological characterization,the coefficients of low-and high-frequency subbands were integrated using targeted fusion rules.Finally,the fused geophysical data were determined through inverse wavelet transform.[Results and Conclusions]Compared to data from a single source,the fused data incorporated the characteristics of structural boundaries delineated using the in-seam seismic exploration,supplemented by information on underground electrical contrast obtained using the electrical method.Therefore,these data allow for the presentation of more abund-ant details of geological structure characteristics.The proposed data fusion technology holds broad application prospects in fields such as geologic hazard prediction and resource exploration,providing a novel,effective approach for the devel-opment of geophysical exploration technology.关键词
数据融合/小波变换/特征识别/主成分分析/槽波/瞬变电磁/音频电透视Key words
data fusion/wavelet transform/feature identification/principal component analysis(PCA)/in-seam wave/transient electromagnetics/audio-frequency electrical penetration分类
天文与地球科学引用本文复制引用
韩姗珊,张继锋,范涛,刘强,宋晓蛟,戚志鹏,郑航宙..多特征联合决策的矿井数据融合技术[J].煤田地质与勘探,2026,54(4):216-226,11.基金项目
国家自然科学基金项目(42174168) （42174168）
陕西省秦创原"科学家+工程师"队伍建设项目(2025QCY-KXJ-024) （2025QCY-KXJ-024）
