Coal mine bumps have presented serious mining problems in the United States throughout the 20th century. Fatalities and injuries have resulted when these destructive events occurred at the working faces of the mines. Persistent bump problem not only threaten the safety of miners, but also have caused the abandonment of large coal reserves and have led to premature mine closures. This research project focuses on coal bumps prevention in long-wall coalmines. A new method for gate-road design using yield pillar in long-wall coalmines will be developed for bumps and floor heave control. Current practice of yield pillars application shows that it can prevent coal bumps in some long-wall mines. The proposed approach will provide a method for stable yield pillar design. To differentiate between stable and violent failure of rocks, an approach proposed by Crouch and Fairhurst, (1) and Salamon (2) will be used in this method. The technical approach designed to meet this objective comprises three phases. In Phase 1, a survey of mining practices and problems related to coal bumps in long-wall coal mines will be conducted. Also, the most significant factors that contribute to coal bumps, such as geological, geometrical, geomechanical and mining variables will be defined statistically. In Phase 2, the main objective of this phase is to simulate the performance of yield pillar system at different stages of long-wall mining using large 3-D finite element models. These models will only consider the most significant variables obtained in Phase 1. Modeling results will be verified by field measurements. These measurements will include changes in pillar stress, pillar dilation, roof-to-floor closure, etc. In Phase 3, using the FE modeling results obtained in Phase 2, a multi-regression model correlating the pillar size to studied variables will be developed. This model will be adjusted and verified using actual cases of mines using yield pillar systems.
