Abstract: The rating of rockburst is fuzzy and uncertainty, though the cloud model of rough set theory has a unique advantage to dealing with fuzzy and uncertainty problem. A new cloud model for evaluating rockburst grade is proposed on the basis of the rough set of fuzzy C-means (FCM) algorithm. Some main evaluation factors of rockburst are chosen in this study, such as uniaxial compressive strength , the maximum tangential stress , tensile strength and elastic energy index Wet . According to the criteria of rockburst classification, each evaluation factor is calculated by using the characteristics of cloud number of different rockburst levels. The rough set theory based on FCM algorithm is employed to analyze each factor attributed in the forty samples of rockburst engineering in China and internationally, and then the weight of each evaluation factor is calculated. The comprehensive degrees of certainty of the samples for evaluation are calculated by the positive normal cloud generator, and the level of rockburst is finally specified by the maximum certainty degree. The results show that the evaluated results agree well with the practical records, which implies that the proposed procedure is of feasibility and reliability, and is a new idea for rockburst prediction.

Key words: rating of rockburst, cloud model, rough set, fuzzy C-means algorithm, comprehensive degrees of certainty