An interval fuzzy comprehensive assessment method for rock burst in underground caverns and its engineering application

Rock burst is a dynamic process involving the sudden release of elastic energy accumulated in overstressed rocks and coal masses during underground excavations, capable of causing a great number of fatalities, failure of supporting structures, and damage to equipment. To minimize the risk, a fuzzy comprehensive assessment model was established to predict and evaluate rock burst in underground caverns, and some indices were selected and analyzed. The result vectors and weight vectors were expressed by interval numbers and the membership degrees were determined by a sigmoid membership function. In addition, synthetic weight was obtained by combining subjective weight with objective weight, and final vectors were categorized into different levels through possibility ranking analysis. Finally, the proposed method was applied to a practical case, the Jiangbian hydropower station in China, for further verification. Considering that some indices were conceptually dependent on one another, the multicollinearity of these indices was evaluated according to results based on different indicator systems. The evaluation results showed a high consistency with the actual situation. Moreover, the indicator system, which made full use of the obtained data, had more reliable results than other indicator systems. The proposed fuzzy comprehensive assessment method provides valuable guidance for rock burst assessment.