论文部分内容阅读
提出了边坡应力分析的新方法，即离散元计算与神经网络预测相结合的模型。首先应用离散元计算不同的开采深度下的边坡岩体应力，将计算结果形成样本集，应用神经网络学习并进行推理，预测其他开采深度的边坡岩体应力。经已开挖的边坡岩体应力验证，该分析模型既方便又合理，并能进一步推广用于预测未开挖的边坡岩体应力。
A new method of slope stress analysis is proposed, which is a combination of discrete element calculation and neural network prediction. Firstly, the discrete elements are used to calculate the stress of slope rock mass under different mining depths. The calculation results are formed into a sample set. The neural network is used to study and carry on the reasoning to predict the stress of the slope rock mass at other mining depths. After the stress analysis of the excavated slope rock mass, the analysis model is convenient and reasonable, and can be further popularized for predicting the stress of the rock mass without excavation.