Relationship between Surface Subsidence Range and Geological Mining Conditions Using Numerical Simulation and Machine Learning

With the increase of mined-out areas and geological disasters, it is necessary to study influence range mining. Moreover, with growth in amount data, machine learning methods are commonly used predict surface subsidence. The 10 mm subsidence boundary an important indicator determine range, FLAC3D suitable for solving nonlinear large deformation problems. Therefore, can be explore relationship between movement (L) mining conditions. Taking working faces A B two mines, through establishment a numerical model, L bedrock thickness, loose layer face length, overlying rock lithology explored by using control variable method. variation law analyzed theoretically mechanically. results show that positively correlated while negatively length. Similarly, shows slight after key becomes weak. This provides scientific basis reasonable coal protection buildings area. Integrating into model further increases throughput. findings models had high accuracy predicting subsidence, random forest trees regression having lower time accuracy, respectively.