Probability prediction method for rockburst intensity based on rough set and multidimensional cloud model uncertainty reasoning

Abstract

Rockburst is a serious disaster caused by the sudden release of rock energy during underground construction in high-stress environments, resulting in severe damage to underground structures. Accurately predicting rockburst intensity is challenging, and establishing a reliable and precise prediction model is of great importance. In this study, we proposed a novel hybrid model for predicting rockburst intensity by integrating rough set theory and multidimensional cloud model uncertainty reasoning. The key steps of the proposed method are as follows: (1) Rockburst cases are collected, and the maximum shear stress ${\sigma }_{\theta }$, uniaxial compressive strength ${\sigma }_{c}$, uniaxial tensile strength ${\sigma }_{t}$, and elastic energy index ${W}_{{\text{et}}}$ are used as predictors for rockburst strength. (2) The Shannon entropy method is used to determine the weights of the four indicators, and a rockburst potential expression is constructed. (3) Rough set theory is used to reduce the number of indicators to construct a rockburst strength prediction rule library. (4) Qualitative data are transformed into quantitative data using the rules library and multidimensional cloud model to establish an uncertainty inference framework for predicting rockburst strength. Finally, we compared the performance of the hybrid model with existing models, and the results demonstrate that the proposed approach achieves similar or even higher prediction accuracy. The use of cloud droplets in the model offers a significant advantage in the prediction of mixed rockburst intensities, enabling intuitive, rapid, and effective determination of the occurrence intensity of rockburst.