Research on In Situ Stress Measurement and Inversion, and its Influence on Roadway Layout in Coal Mine with Thick Coal Seam and Large Mining Height

Seven points were selected to measure in situ stress, and the simulation software ANSYS was used to introduce a 3D geological model of Licun colliery, which has a thick coal seam and large mining height. A BP neural network was applied to the inversion of in situ stress, and distributions of σ₁, σ₂, and σ₃ were achieved. This proved that a BP neural network can be applied to the inversion analysis of the colliery’s in situ stress. Furthermore, different angles α between the maximum principal stress and roadway axial were simulated. The results showed that as α increases, the shallow maximum principal stress varies from symmetric distribution to asymmetrical distribution. Finally, this becomes a symmetric distribution as the concentrated contours vary from the roof to two sides at the same time. The maximum tensile stress of the roof and α were positively correlated, and the depth of the maximum tensile stress and α were negatively correlated. The tensile stress of the roof surrounding shallow rock increases as α increases, and the roof tends to cause tensile failure. Thus, the roadway axial and the maximum principal stress should be laid out in parallel as much as possible, and angle α should be less than 30º in the Licun colliery.