Rock burst risk assessment in deep-buried underground caverns: a novel analysis method

Abstract

A rock burst often occurs during the construction period of deep-buried underground caverns. How to predict and prevent it is an urgent problem in underground engineering, especially in large hydropower stations. Combining grey correlation method, principal component analysis (PCA) and cloud theory, a novel analysis method that is proposed to evaluate rock burst. First, seven indices, namely, Rc, Rc/σ1, Rc/σt, σθ/Rc, Wet, H and KV, are selected. Considering the relationship between these indices, the grey correlation method is used to analyze these indices and reduce them. According to the correlation coefficients, the five indices, namely, Rc/σ1, Rc/σt, σθ/Rc, Wet and KV, consist of the final evaluated system. Second, the weight of each index is calculated using principal component analysis (PCA). Take into consideration of the ambiguity and randomness of rock burst; the multi-dimensional cloud model is used to evaluate the rock burst level. The proposed model is applied to a case study of the Jiangbian hydropower station to certify the feasibility and effectiveness of the novel method. The results are basically consistent with the actual rock burst level. At last, the selection of the evaluation system and the accuracy of the multi-dimensional cloud model are discussed. This novel method provides a new idea for the risk assessment of rock burst in deep-buried underground caverns.