摘要
滑坡是一种重要的地质灾害,合理地治理滑坡是工程建设的关键,而合理的力学参数对滑坡治理又起到重要的作用,其中滑坡滑带力学参数是最关键参数之一。根据现场踏勘和地质资料分析,对滑坡滑带参数在含水条件和天然条件下进行反演分析、利用采集的数据为学习样本进行神经网络学习,对室内实验3组数据进行BP神经网络预测土样的C、Φ值,并参照已知滑坡滑带的力学参数来综合确定该滑坡滑带的力学参数,为合理治理该滑坡提供依据。
Landslide is an important geological hazard in nature. It gets great threat for people's life and property. Reason- able landslide treatment is the key of engineering construction. Reasonable mechanical parameter is important for landslide control. The mechanical parameter of landslide zone is one of the most essential parameters. Based on the field reconnaissance and analysis of geological data, it takes back analysis on landslide zone parameters under water condition and natural condition. It also studies the neural network through taking the selected data as studying samples. It uses three groups of experimental data indoor with BP neural network to predict C and Ф values of soil samples. Then it determines the mechanical parameters of this landslide zone comprehensively based on known mechanical parameters of landslide zone. Finally, it provides basis for reasonable treatments of landslides.
出处
《水科学与工程技术》
2009年第3期46-48,共3页
Water Sciences and Engineering Technology
关键词
滑坡滑带参数
反演分析
BP神经网络
综合确定
landslide zone parameter
back analysis
BP neural network
comprehensive determination