CAJ | 학술논문

为了对滑坡的稳定性进行分析，文中采用BP神经网络的方法，通过收集滑坡数据作为研究样本，选取平均坡度、前缘临空坡度、滑面坡度、多年平均最大日降雨量、近期滑动情况、人类工程活动6个因素作为滑坡的影响因素。并应用Matlab工具箱对神经网络进行了设计、权值初始化、数据归一化及网络训练，提出了滑坡稳定性分析模型，对误差进行分析。其结果表明样本训练结果和实际情况基本吻合，证明该模型能够符合工程实际。
위료대활파적은정성진행분석，문중채용BP신경망락적방법，통과수집활파수거작위연구양본，선취평균파도、전연림공파도、활면파도、다년평균최대일강우량、근기활동정황、인류공정활동6개인소작위활파적영향인소。병응용Matlab공구상대신경망락진행료설계、권치초시화、수거귀일화급망락훈련，제출료활파은정성분석모형，대오차진행분석。기결과표명양본훈련결과화실제정황기본문합，증명해모형능구부합공정실제。
Through collecting the landslide data as the research sample and selecting average slope , slope of the invading front , slope of the sliding surface , the maximum annual average rainfall , the recent slide situation and human activities as the influence factors of landslide , BP neural network method is used to analyze the stability of the landslide .Applying Matlab toolbox to design neural networks , weight initialization , data normalization and network training , a landslide stability analysis model is put forward , and the error analysis is conducted .The results show that sample training re-sults agree with the actual situation basically , demonstrating that this model tally with the project re-ality.