SURFACE SUBSIDENCE PREDICTION BY NONLINEAR FINITE-ELEMENT ANALYSIS

An improved two-dimensional plane-strain numerical procedure based on the incremental-iterative nonlinear finite-element method is developed to predict ground subsidence caused by underground mining. The proced ure emphasizes the use of the following features: (1) An appropriate c onstitutive model that can accurately describe the nonlinear behavior of geological strata; and (2) an accurate algorithm for simulation of excavation sequences consistent with the actual underground mining pro cess. The computer code is used to analyze a collapse that occurred in the Blue Goose Lease #1 Mine in northeastern Oklahoma. A parametric s tudy is conducted to investigate the effects of some selected factors on the shape and extent of subsidence profiles. Analyses of the numeri cal results indicate that the nonlinear finite-clement technique can b e employed to meaningfully predict and characterize the potential for ground subsidence due to underground mining.