Coal-Free Zone Genesis and Logging Response Characterization Using a Multi-Curve Signal Analysis Framework.

Coal-free zones, particularly scouring zones, reduce recoverable reserves and increase water inrush risk in coal mining. Existing sedimentological, geophysical, and geostatistical methods are often constrained by coring conditions, lithological interpretation accuracy, and geological complexity. Given that well log signals provide the most continuous carriers of geological information, this study integrates Singular Spectrum Analysis (SSA), Maximum Entropy Spectral Analysis (MESA), and Integrated Prediction Error Filter Analysis (INPEFA) to establish a multi-curve framework for analyzing the genesis and logging responses of coal-free zones. A two-stage SSA workflow was applied for noise reduction, and a Trend-Fluctuation Composite (TFC) curve was constructed to enhance depositional rhythm detection. The minimum singular value order (N), naturally derived from SSA-decomposed INPEFA curves, emerged as a quantitative indicator of mine water inrush risk. The results indicate that coal-free zones resulted from inhibited peat-swamp development followed by fluvial scouring and are characterized by dense inflection points and frequent cyclic fluctuations in TFC curves, together with the absence of low anomalies in natural gamma-ray logs. By integrating multi-curve logs, core data, and in-mine three-dimensional direct-current resistivity surveys, the genetic mechanisms and boundaries of coal-free zones were effectively delineated. The proposed framework enhances logging-based stratigraphic interpretation and provides practical support for working face layout and mine water hazard prevention.