Background The advance detection of underground concealed water hazards in coal mines is crucial to ensuring the safe and efficient roadway tunneling. The borehole transient electromagnetic (TEM) method allows for long-distance advance detection. However, due to the limitation of the spatial structure of boreholes, currently available techniques can only effectively identify low-resistivity anomalies (e.g., water-bearing structures) in the radial direction, leaving blind areas in the primary tunneling direction (i.e., the axial direction of boreholes).

Objective and Method To address this challenge, this study proposed a method for the advance detection of concealed water hazards in the axial direction of boreholes using both the vector synthesis based on biorthogonal transmitter-receiver coils and a self-adaptive inversion algorithm. By designing vertical and horizontal orthogonal transmitter coils, this method excited dual-component primary fields using sequential current while collecting the time-domain signals of secondary fields in the corresponding direction. Based on the vector synthesis principle, data on dual-component anomaly fields extracted using the trend surface method were fused to produce multi-directional composite anomaly fields. The composite anomaly fields were then superimposed on the background field obtained from numerical simulation, yielding the total field with high signal-to-noise ratios (SNRs). This enables the fan-shaped advance detection of concealed water hazards in the axial direction of boreholes. During the inversion stage, a self-adaptive regularization algorithm with physical constraints, combined with the main direction angle of the synthesized field, was employed to achieve the 3D imaging of low-resistivity anomalies.

Results and Conclusions  Both the numerical simulation (achieved by constructing a 3D model with a low-resistivity anomaly using the FDTD method) and the physical simulation of a 1:100-scale flume (obtained by simulating a water-bearing structure using highly conductive media) indicate that the proposed method can accurately locate the anomaly, with the degree of match between imaging morphology and the pre-set model reaching up to 90 %. The proposed method was applied to the detection of a water sump in a horizontal roadway of a coal mine in Shanxi Province. The water hazard target was effectively identified by comparing images captured before and after water discharge. Furthermore, the method was employed to detect the water accumulation range in the goaf of mining face 15101 within the coal mine based on three verification boreholes. The inversion results reveal the presence of a stable low-resistivity anomaly zone in borehole intersection area, confirming the expansion trend of the water accumulation range in the goaf over time. The application results further demonstrate the reliability and engineering applicability of the proposed method in localizing concealed water hazards. The results of this study provide effective technical support for the long-distance, high-precision advance detection of underground concealed water hazards in coal mines, holding critical significance for ensuring the safe coal mining of mines.