论文部分内容阅读
针对线性回归法在地应力拟合精度不高的问题,通过利用神经网络法和线性回归分析法对沁水盆地南部地区的地应力进行预测并分析。结果表明,基于神经网络算法的地应力预测值与实测值拟合更好,并证明了神经网络在地应力预测中的应用更符合工程设计要求。
In view of the problem that the linear regression method is not suitable for the precision of ground stress fitting, the earth stress in the southern Qinshui Basin is predicted and analyzed by using neural network and linear regression analysis. The results show that the predicted value of in-situ stress based on neural network algorithm fits well with the measured value, and proves that the application of neural network in the prediction of ground stress is more in line with the engineering design requirements.