边坡稳定性分析中工程岩体抗剪强度参数选取的神经网络方法  

ARTIFICIAL NEURAL NETWORK METHOD TO PREDICT SHEAR STRENGTH OF ROCKMASS FOR ANALYSIS OF SLOPE STABILITY

作　　者：许传华[1] 房定旺[2] 朱绳武[2] 

机构地区：[1]河海大学土木工程学院,南京210098 [2]马鞍山矿山研究院,马鞍山243004

出　　处：《岩石力学与工程学报》2002年第6期858-862,共5页Chinese Journal of Rock Mechanics and Engineering

摘　　要：影响工程岩体抗剪强度参数的因素很复杂,具有不确定性。由于神经网络方法不仅能考虑定量因素,而且能考虑定性因素的影响,因而神经网络适用于解决非确定性的工程岩体抗剪强度参数的选取问题。建立了选取工程岩体抗剪强度参数的人工神经网络模型,并利用大量工程资料对网络进行训练和测试,得出了一些有意义的结论。The factors which control and affect the shear strength of rockmass are complicated and uncertain variables. As artificial neural network method can consider both quantitative and qualitative factors,it is suitable to solve the uncertain problems with artificial neural network method. The neural network model is establlished for prediction of shear strength of engineering rockmass and a lot of engineering data are collected to train and examine the model. Some valuable results are derived at last.

关 键 词：工程岩体抗剪强度 神经网络 边坡稳定性 

分 类 号：TU457[建筑科学—岩土工程] TP183[建筑科学—土工工程]