首页 | 本学科首页   官方微博 | 高级检索  
     检索      

边坡稳定性预测的模糊神经网络模型
引用本文:薛新华,姚晓东.边坡稳定性预测的模糊神经网络模型[J].力学学报,2007,15(1):77-82.
作者姓名:薛新华  姚晓东
作者单位:1. 浙江大学岩土工程研究所,杭州310027; 2. 中天建设集团,杭州310008
基金项目:国家自然科学研究基金(编号50379046)资助研究项目
摘    要:根据边坡稳定问题具有的模糊性,提出了一种判定边坡稳定性的模糊神经网络模型。该系统仅从期望输入输出数据集即可达到获取知识、确定模糊初始规则基的目的。再利用神经网络学习能力便不难修改规则库中的模糊规则以及隶属函数和网络权值等参数,这样大大减少了规则匹配过程,加快了推理速度,从而极大程度地提高了系统的自适应能力。最后用收集到的边坡数据样本训练和测试模糊神经网络模型,结果表明该模糊神经网络预测边坡稳定性是可行的、有效的。

关 键 词:混合学习算法  边坡稳定  神经网络  模糊性
收稿时间:2006-04-12
修稿时间:2006-07-06

A FUZZY NEURAL NETWORK MODEL FOR PREDICTING SLOPE STABILITY
XUE Xinhua,YAO Xiaodong.A FUZZY NEURAL NETWORK MODEL FOR PREDICTING SLOPE STABILITY[J].chinese journal of theoretical and applied mechanics,2007,15(1):77-82.
Authors:XUE Xinhua  YAO Xiaodong
Institution:1. Institute of Geotechnical Engineering, Zhejiang University, Hangzhou 310027; 2. Zhong Tian Construction Group, Hangzhou 310008
Abstract:Accordingto the fuzzy characteristics of slope stability,a fuzzy neural network model is presented to predict slope stability.In this model,theintention of acquiringthe initialfuzzy rule sets can be achieved only by using the desired input-output data pairs.Then,employing neural networks learning techniques,the fuzzy logic rules,input-output fuzzy membership functions and weights in network can be easily tuned.So the rule matching is reduced.The velocity of inference is accelerated.Adaptabilityof the system is greatly improved.At last,the collected data of slope stability are adapted to train and test the model.The forecasted results show that the proposed method is feasible and effective in predicting slope stability.
Keywords:Hybridtraining algorithm  Slope stability  Neural network  Fuzziness
点击此处可从《力学学报》浏览原始摘要信息
点击此处可从《力学学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号