关键词: 粗糙集理论, 岩爆预测, 分级, 知识约简

Abstract: Based on the combination of rough set and fuzzy set theory，a knowledge reduction method of rough set is adopted to obtain the evaluated indexes.The problem of weighting coefficient is converted into that of significance estimating on the attribution of rough set.And a model of predicting the possibility and classification of rock burst was established.A series of underground rock projects in domestic and abroad are taken as the training and testing samples，and the ratio of mistake-distinguish is considerably low zero after the model is trained.Ma Luping deep-buried hard rock mine are used to verify this model.The results show that the predicted classification is consistent with the actually measured ones.Therefore，the classification model，reflecting the complexity situation of rock-burst objectively，is effective and available to the classification of the grade of rock burst.The results show that the model can be used to predict the possibility and classification of rock-burst in deep underground engineering.

Key words: Rough sets, Rock-burst prediction, Classification, Knowledge reduction