Evaluation of Rock Burst Tendency in Western Hubei Based on BP Neural Network

Abstract Rock burst tendency evaluation is the basis of rock burst risk assessment in underground engineering. In order to realize the accurate evaluation of rock burst tendency in western Hubei, the four parameters of rock burst strength, uniaxial compressive strength of rock σc, the ratio of uniaxial compressive strength to tensile strength σc/σt, rock mass integrity coefficient Kv, the ratio of tangential stress to uniaxial compressive strength σθ/σc were selected as the evaluation index of rock burst tendency in western Hubei.In order to realize the measurement of the above indexes, the location and phenomenon of the rock burst in western Hubei are summarized, and the rock burst is sampled. Then, through the combination of the field and indoor tests, the measurement of the rock burst tendency evaluation index in western Hubei is completed. Finally, based on the Matlab toolbox, round and rank functions are used to generate random training samples and construct the BP neural network model, and the model was tested. The prediction results show that the rock explosion mainly occurs in dolomite and phosphorus rock, and the evaluation results are consistent with the actual situation. The model is practical and relatively easy to calculate, which reduces the deviation caused by individual influencing factors and is easy to popularize in practice.