Drilling-parameter-based identification of potential roof collapse zones in coal mine roadways: laboratory investigation and field validation

Early identification of potential roof collapse zones in coal mine roadways and implementation of effective prevention and control measures is of important practical significance for reducing the probability of roof fall accidents and preventing casualties. By investigating the sensitivity response relationship between drilling signals and the strength, thickness and assemblage of roof strata during the drilling process, this study reveals the attenuation law of vibration parameters and cutting parameters under drilled rock mass deterioration, and develops a method for identifying roof strata and potential roof collapse zones based on drilling parameters. The results show that drilling vibration and pressure responses are strongly controlled by specimen strength and integrity. From high- to low-strength specimens, the average vibration acceleration decreased from 1.367 to 0.865 g. The root mean square value decreased from 1.746 to 1.211 g, and the centroid amplitude decreased from 2578.294 to 534.828. The average pressure decreased from 12.954 to 7.387 kN, while the peak density decreased from 4.043 to 1.447 mm −1 . Compared with intact specimens, fractured specimens showed clear attenuation in drilling vibration response, with attenuation rates of key vibration parameters ranging from 30.47 to 82.74%. This method realized the real-time acquisition of drilling information and the accurate calibration of the three-dimensional shape of the potential caving area of roadway roof. Besides, when the drill bit crossed rock-layer interfaces or separation zones, vibration acceleration dropped sharply and approached zero, providing a distinct indicator for interlayer discontinuities. Field monitoring at Liuwan Coal Mine further confirmed that 8 m roof boreholes could be divided into five intervals according to vibration-response variations, allowing fractured zones, weak interfaces, and relatively intact layers to be distinguished. Overall, the proposed method provides a practical basis for roof stability evaluation, support-parameter optimisation, and prevention of mining-induced roof instability.