The Rock Burst Hazard Evaluation Using Statistical Learning Approaches

The great threat and destructiveness brought by a rock burst make its prediction and prevention crucial in engineering. The rock burst hazard evaluation at project locations is an effective way of preventing rock burst since currently real-time prediction is not available. Since different control factors and discrimination conditions of rock burst were accepted by conventional risk determination methods, the rock burst risk determination in the same area may produce conflicting results. In this study, Naive Bayes statistical learning models based on different model prior distributions representing highly complicated nonlinear relationship between rock burst hazard and impact factors were built to evaluate the rock burst hazards. The results suggested that the Bayes statistical learning model based on a Gaussian prior has the strongest performance over four preset prior distributions. Combining the rock mechanics parameters measured in the laboratory and the stress data collected on the project sites, the proposed model was successfully employed to evaluate the kimberlite rock burst risk of a diamond mine in Canada. The Bayes statistical learning model exhibits its robustness and generalization in rock burst hazard evaluation, which can be generalized for similar engineering cases with enough supported data.

The results of rock burst prediction on the sokolovskaya underground mining by the acoustic emission method

The relevance of research. When developing underground mining, there is a risk of dynamic phenomenon of high rock pressure - rock burst. The local instrumental forecast of shock hazard by acoustic emission with the GS-01 device allows you to quickly identify high-tension zones during underground mining. The Research aim: generalization of the accumulated data of instrumental measurements of acoustic emission parameters throughout the Sokolovskaya mine; zoning of rock burst hazard in the deposit; use of zoning of rock burst hazard results for mining planning in terms of changing the principles and regulations for designing traces of mine workings in potentially impact hazardous areas; improving the methodology of instrumental forecasting of shock hazard by the acoustic emission method. Methods of research: analysis of the accumulated data of instrumental measurements of acoustic emission parameters during mining; theoretical studies (identifying patterns of changes in acoustic emission taking into account the influence of mining, the configuration of the excavation, etc.); measurement in situ (measurement of acoustic emission parameters when changing parameters underground working). Results of research: implementation of preventive measures (the so-called “relaxation” of the rock) to reduce the rock burst risk in hazard zones during mining; implementation of research results in normative and technical documentation.

Study on Rule of Overburden Failure and Rock Burst Hazard under Repeated Mining in Fully Mechanized Top-Coal Caving Face with Hard Roof

The dynamic disasters caused by the failure of hard roof in the process of mining coal seriously affect the safe production in coal mines. Based on the W1123 mining coal working face of Kuangou coal mine, the physical similar material simulation experiment and acoustic emission (AE) monitoring method are used to study the failure law and AE characteristics of overburden in the process of coal mining. The stress evolution law is revealed through numerical simulation, the dangerous areas and rock burst hazard under the repeated mining with hard roof are studied combined with microseismic monitoring on site. The results show that the energy of W1123 working face released by the overburden damage under B4-1 solid coal is higher than that of the gob, and the peak value of the AE energy appears near the W1145 open-off cut. Through the statistics of the AE data, the large energy rate of AE event is defined, and the AE events with large energy rate appear in the scale of 82.4–231.2 cm within the model. This area is shown as a stress superposition area according to the numerical simulation. On the basis of comparing with the characteristics of energy distribution in the field, it is considered that the main control factors of rock burst in this area are hard roof of the working face and the stress concentration caused by the repeated mining. It provides a scientific guidance for the prevention and control measures of rock burst in this type of mining condition.