金川矿区地应力规律与人工神经网络预测研究

LAW OF IN-SITU STRESS IN JINCHUAN DIGGINGS AND RESEARCH ON PREDICTION MODEL BASED ON ARTIFICIAL NEURAL NETWORK

介绍了对金川矿区地应力变化规律的研究与预测结果,在矿区地应力测试资料的基础上,开发了矿区地应力神经网络预测模型。借助于该模型,进行矿区三维空间的地应力预测和分析,给出了由此获得的主要结论。

The test result of in-situ stress in Jinchuan diggings demonstrates that the area is of higher in-situ stresses. The destruction model and stability of the tunnel largely depend on the quantity and direction of in-situ stress. As the excavation depth and in-situ stress increase,the manipulation of the field stress in the excavation area becomes harder. Thus,the disclosure of the in-situ stress law is important for optimum control technology and tunnel supporting design. The test material of in-situ stress in the diggings is collected,and the prediction model of in-situ stress is developed,based on neural network theory. The main conclusion is achieved after the prediction and analysis of in-situ stress in the digging.