滑带土强度参数的人工神经网络预测

长江科学院院报 ›› 2003, Vol. 20 ›› Issue (2): 34-36.

滑带土强度参数的人工神经网络预测

王世梅, 刘德富, 田斌

出版日期:2003-04-01 发布日期:2012-03-05

Neural network prediction for strength parameters of landslide zone soil

WANG Shi-Mei, LIU De-Fu, TIAN Bin

Online:2003-04-01 Published:2012-03-05

摘要/Abstract

摘要： 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。

Abstract: Using the improved BP net technique, a new methd fordetermination of strength parameters of landslide zone soil is presented. Based on analysis of some factors affecting strength parameters, taking data of some landslides as training and testing samples, trained by momentum method, the prediction model of stregnth parameters is prososed. The results show that the model has a high precision, this method has been successfully used in calculating strength parameters of Gushubao landslide zone soil, and obtained results having good agreement with those by other common methods.

王世梅, 刘德富, 田斌. 滑带土强度参数的人工神经网络预测[J]. 长江科学院院报, 2003, 20(2): 34-36.

WANG Shi-Mei, LIU De-Fu, TIAN Bin. Neural network prediction for strength parameters of landslide zone soil[J]. JOURNAL OF YANGTZE RIVER SCIENTIFIC RESEARCH INSTI, 2003, 20(2): 34-36.