Abstract: Rockburst is one of the main geological disasters in underground excavation and the classification prediction of its intensity is a worldwide problem that needs to be solved urgently. In view of the uncertainty in prediction, the rock shear stress to uniaxial compressive strength ratio σ_θ/σ_c, the rock uniaxial compressive strength to tension strength ratio σ_c/σ_t and elastic energy index W_et were selected to define the rockburst evaluation indexes. The entropy weight combined with improved CRITIC method was adopted to determine the weighting coefficient of each evaluation index. Combined with the theory of artificial intelligence with uncertainty, the algorithm of backward cloud generator was used to establish 3 digital features of the multi-dimensional cloud model and generate the multi-dimensional cloud model including all the prediction indicators. Finally, the accuracy and validity of the proposed model were validated with case data of 48 groups of typical rockburst both at home and abroad. Furthermore, results obtained by the proposed model were compared with those got by cloud model based on weighted fusion and one-dimensional cloud model. The results show that the proposed model has higher accuracy in rock burst prediction.

Key words: rock mechanics, rockburst prediction, multidimensional cloud model, entropy weight, algorithm of reverse cloud generator, criteria importance though intercriteria correlation(CRITIC)