Prediction Of Rockburst While Tunnelling In High Geological Stress Areas

Abstract:Rockburst is a kind of dynamic geological disaster happened in high stress areas when mining and tunnelling in hard and complete rock. Generally, it is caused with a complex mechanism and usually breaks out abruptly and randomly, which always brings huge damage to the safety of life and property for people in a very short time. If it can’t be predicted in time, it will not only damage people and equipment, but also influence the construction schedule, which will result in serious consequences such as the increase of comprehensive cost of the construction. Therefore, in order to make the design and construction more reasonable, effective and economical, this project mainly study the rockburst prediction theory on the ground of Qinling tunnel.1) The influence factors of rockburst are analyzed based on the rockburst mechanism, including the maximum tangential stress, rock uniaxial pressure strength, rock uniaxial tensile strength and elastic energy index. Then, the ratio of the maximum tangential stress, the ratio of rock uniaxial pressure strength and rock uniaxial tensile strength, and elastic energy index are chosen as indexes to evaluate rockburst. The importance of each index is studied. According to the existed relative research, taken these three indexes into consideration the index multifactorial evaluation system of rockburst is established.2) Ideal point model for prediction of rockburst is built based on those indexes. A combination weighting method which combines improved analytic hierarchy process (AHP) with entropy value method is used to evaluate the importance of different indexes in this model. In addition, Euclidean distance is adopted to reflect the positive and negative ideal point distance of rockburst samples under different rockburst intensity levels. It is thought that the rockburst intensity level is higher if the sample is closer to the positive ideal point. Moreover, the model is applied to practical projects such as Dongguashan copper mine and Lingbaodongyu mining area to predict the rockburst, and results show that this model is reasonable.3) Variable weight target degree model of rockburst prediction is established. In this model, an interval correlation function is constructed to define attribute values of different indexes under different rockburst intensity levels. And targets of different rockburst intensity levels are calculated according to these attribute values. Moreover, both the analytic hierarchy process (AHP) and a kind of equilibrium function are used in this model to give weights to the indexes, which can make data of the indexes be more reasonable and impersonal. In addition, approach degree of different rockburst intensity levels can be obtained by calculating the distance between samples and targets. At last, the model is applied to practical projects such as Dongguashan copper mine and Lingbaodongyu mining area to predict the rockburst, and results show that this model is reasonable and useful.4) Rockburst prediction model system is developed. In the actual use of this system, users can straightly predict rockburst intensity after guiding the data into the system. In order to verify the reasonability of this model, it is applied to rockburst prediction of Qinling tunnel combined with numerical methods. Results are coincident with the actual situations, indicating that this model is reasonable and useful. To be more exact, mild to moderate rockburst will happen in the roof and the bottom and their near sides of the tunnel.