The Fifth International Symposium on Mine Safety Science and Engineering (ISMSSE2021)

29 / 2020-03-12 13:58:34

Back Analysis for Initial Ground Stress Field at a Diamond Mine Using Machine Learning Approaches

Initial ground stress field; Full-scale finite element model; Multi-output decision tree regressor; Neural network

Exact knowledge for ground stress field guarantees the construction of various underground engineering projects. Limited by costs, field measurement for initial ground stresses can be only conducted on several measure points, which necessitates back analysis for initial stresses from limited field measurement data. This paper employed multi-output decision tree regressor (DTR) to represent the relationship between initial ground stress field and its impact factor. A full-scale finite element model was built and computed to gain 400 training samples for DTR with a sub-modelling strategy. The results showed that correlation coefficient r between field measurement values and back analysis values reached 0.9524, which proved the success of DTR. A neural network was employed to reflect a global initial ground stress field. More than 600,000 node data extracted from the full-scale finite element model were used to train this neural network. After training, the stresses on any location can be obtained from this model with the inputs of corresponding coordinates.