Stochastic distribution characteristics of void fraction in overburden disturbed by longwall mining in shallow seam

Multi-scale voids including fractures and bed separations will appear and dynamically develop as the underground mining-induced disturbance increases. The void fraction, which is an index for characterizing the void scale, is an important parameter for understanding the heat and mass transfer (including methane flow, coal spontaneous combustion, underground water flow) in the fractured overburden disturbed by underground mining of coal seam. The stochastic caving tests were conducted to simulate the roof caving and overburden movement induced by coal seam mining with different inclination angle and excavation length. The expectations and variances of random variables about the overall and local void fractions showed that the void fractions gradually increase with increases in excavation lengths and obey a normal distribution. The distribution fields of the mean and stochastic void fractions indicated that the rich voids present an arch-type distribution in the disturbed overburden. The random distribution field of void fractions obtained from stochastic inversion presents a small range of fluctuation around the mean values of void fractions and is more realistic to reflect the randomness of separation and caving of rock strata induced by underground mining of coal seam.