Research on rockburst prediction with fuzzy comprehensive evaluations based on rough set

Being a familiar geological hazard in underground engineering, rockbursts greatly threaten the safety of personnel and equipment. A rockburst prediction method was proposed using fuzzy comprehensive evaluations based on rough sets. A method of determining weight coefficients based on the rough set theory is presented in this paper. Determining much weight coefficients was translated into estimating significance of attributes among rough sets. Based on the analysis of main causes of rockbursts, the compressive strength, tensile strength, the included angle of preferred structural planes, maximum principal stress, elastic energy index of rock, and the maximum tangential stress of the tunnel wall were chosen as the indexes. The decision table for rockburst prediction was established. By analyzing the significance of conditional attribute sets, the weight coefficient of fuzzy comprehensive evaluations model was determined. These weight coefficients are determined completely from data, which overcomes the subjectivity of experts. Finally, the model was applied to predict the rockbrust gradation of the Kuocangshan tunnel.