A cloud model for predicting rockburst intensity grade based on index distance and uncertainty measure

Zhang Biao,Dai Xing-guo

DOI: https://doi.org/10.16285/j.rsm.2017.S2.036

2017-01-01

Rock and soil mechanics

Abstract:Rockburst disaster prediction is still an unsolvable global problem in underground rock engineering construction at present; to predict the rockburst intensity grade, a prediction method of rockburst intensity classification of finite intervals cloud model based on index distance and uncertainty measure is proposed to overcome the measured index values of certainty and uncertainty, intensity classification are of fuzzy and random in predicting. The evaluation index system of rockburst intensity classification forecasting is first established based on a comprehensive analysis of the rockburst mechanism and the Mass function of each index is calculated by the ridge function, the weight coefficients are second obtained based on the index distances and uncertainty measure. Then the normal cloud theory is modified and used to calculate the cloud characteristics for each evaluation index in rockburst classification, which generates the cloud drops in finite intervals, combined the measured index values with the corresponding weights the comprehensive degrees of certainty are obtained; and the rockburst level is identified by the weighted average principle in the end; the uncertainty and randomness mapping between the semantic variable and the evaluation index value are realized. The actual cases are introduced to further explain the calculation flow of the prediction model; and comparing with other theory methods shows that the model proposed is effective for rockburst classification to a certain extent; and its accuracy is higher than the other methods, so as to provide a novel idea for similar engineering problems.