Prediction of rock failures in mines, with application to the Näsliden mine in northern Sweden

Abstract: Continous cut and fill mining will cause increased loading of the roof of the stopes in mining operations. The aim of the present work is to predict what mining levels will induce critically large loads in the roof of the stopes in cut and fill mining and in particular the Näsliden Mine, in Northern Sweden. The quasi-static elastic response of rock masses in the Näsliden Mine is predicted in terms of the stresses and strains induced by mining operations, by using the Finite Element Method. Critical stages in the development of failures are identified and defined from observations in the mine. In-situ measurements are also taken from a selected reference stope in the mine, at certain selected stages in a sequence of mining operations. The extension strain criterion of failure is used in our calculations. From our calculations, and the observations from the reference stope, critical levels of mining, for all stopes in the mine, are identified. The predicted response of the rock mass from calculations based on a linearly elastic model is compared with predictions from calculations based on a joint element model. The joint elements are used to simulate weak contact adjacent to the ore body. Both models are evaluated and adjusted by comparisons with in-situ measurements of displacements and stresses. A sensitivity study is used to investigate the effect of variation in rock properties and calculation errors due to numerical discretizations. Full scale strength data are compared with data from uniaxial compression tests conducted in the laboratory. The results obtained with the elastic model and the extension strain criterion show satisfactory agreement with data obtained from the mine. Almost identical results, as far as the prediction of roof failure of stopes in the mine is concerned, are obtained when the joint element model and the Coulomb failure criteria is used in our calculations. Our quantitative prediction of rock failure can be used to determine when rock support is needed or when changes in mining methods should be adopted. Our computational model could also be used to investigate alternative mine layouts.