论文部分内容阅读
利用改进BP网络技术,提出了确定滑带土强度参数的新方法。分析了滑带土强度参数的影响因素,并以收集的部分滑坡的实际数据作为训练和模型检验样本,采用动量法进行训练建立预测模型。模型检验表明,训练的模型具有较高的精度。该方法在确定清江水布垭古树包滑坡滑带土强度参数中效果较好,所得数据与其它常规方法综合确定的参数基本一致。
Using improved BP network technology, a new method to determine the strength parameters of sliding zone soil is proposed. The influencing factors of soil strength parameters are analyzed. The actual data of partial landslide is used as training and model test samples, and the momentum model is used to train the prediction model. The model test shows that the training model has higher accuracy. The method is effective in determining the strength parameters of the Shushuabao landslide in the Qingjiang River. The obtained data are consistent with those determined by other conventional methods.