基于径向基网络模型的面板堆石坝坝顶沉降量预测
|
摘要
随着国家在水利基础建设方面的大力投入以及面板堆石坝自身的优越性,近几年该坝型逐步向超高坝型发展,同时坝体变形的预测也面临着诸多困难。本文提出了一种基于径向基(RBF)网络的面板堆石坝的变形预测模型。该模型充分利用了径向基网络的非线性映射能力,利用收集的历史样本信息,即可预测出面板堆石坝沉降。以水布垭面板堆石坝为例,预测得到的竣工期和满蓄5年后的沉降位移分别为2.156m和2.491m,与实测位移基本一致,相对误差分别为0.748%和0.400%。结果表明预测位移在设计允许范围之内,RBF网络模型具有建模速度快、预测精度高的特点。
With the vigorous investment getting involved in the field of water conservancy infrastructure construction in China,and the superiority of concrete face rockfill dams(CFRD),the ultra-high CFRD dams are constructed gradually in recent years.At the same time,the prediction of the crest settlement of CFRD is confronted with many difficulties.This paper proposed a model for deformation prediction of CFRD based on RBF(RBF)networks.The model makes full use of the nonlinear mapping ability of the RBF networks,using the collected information about the history samples,to predict the settlement of CFRDs.Take Shuibuya CFRD as an example,the predicted settlement values by RBF during completion period and storage after 5years of reservoir fully filled are 2.156 mand 2.491 m,respectively.The predicted results are in accordance with the measured ones,and the relative error is 0.748% and 0.400%,respectively.The results demonstrate that the dam deformation is within a reasonable range,and RBF network model has the characteristics of high speed and accuracy of modeling.
With the vigorous investment getting involved in the field of water conservancy infrastructure construction in China,and the superiority of concrete face rockfill dams(CFRD),the ultra-high CFRD dams are constructed gradually in recent years.At the same time,the prediction of the crest settlement of CFRD is confronted with many difficulties.This paper proposed a model for deformation prediction of CFRD based on RBF(RBF)networks.The model makes full use of the nonlinear mapping ability of the RBF networks,using the collected information about the history samples,to predict the settlement of CFRDs.Take Shuibuya CFRD as an example,the predicted settlement values by RBF during completion period and storage after 5years of reservoir fully filled are 2.156 mand 2.491 m,respectively.The predicted results are in accordance with the measured ones,and the relative error is 0.748% and 0.400%,respectively.The results demonstrate that the dam deformation is within a reasonable range,and RBF network model has the characteristics of high speed and accuracy of modeling.
引文
[1]杨启贵,刘宁,孙役,熊泽斌,等.水布垭面板堆石坝筑坝技术[M].北京:中国水利水电出版社,2010.
[2]杨泽艳,周建平,蒋国澄,孙永娟.中国混凝土面板堆石坝的发展[J].水力发电,2011,37(2):18-23.
[3]Behnia D,Ahangari K,Noorzad A,Moeinossadat SR.Predicting crest settlement in concrete face rockfill dams using adaptive neuro-fuzzy inference system and gene expression programming intelligent methods[J].Journal of Zhejiang University SCIENCE A,2013,14(8):589-602.
[4]Zhou W,Hua J,Chang X,Zhou C.Settlement analysis of the Shuibuya concrete-face rockfill dam[J].Computers and Geotechnics,2011,38:269-80.
[5]Seo M-W,Ha IS,Kim Y-S,Olson SM.Behavior of concrete-faced rockfill dams during initial impoundment[J].Journal of Geotechnical and Geoenvironmental Engineering,2009,135:1070-81.
[6]Kim Y-S,Kim B-T.Prediction of relative crest settlement of concrete-faced rockfill dams analyzed using an artificial neural network model[J].Computers and Geotechnics,2008,35:313-322.
[7]周伟,花俊杰,常晓林,杨启贵,马刚.水布垭高面板堆石坝运行期工作性态评价及变形预测[J].岩土工程学报,2011,33(S1):65-70.
[8]詹振彪,罗先启.BP网络与遗传算法在水布垭工程中的应用[J].岩石力学与工程学报,2002,21(7):963-967.
[9]曹克明,汪易森.高混凝土面板堆石坝的设计与施工[J].水力发电,2001(10):49-52.
[10]Clements RP.Post-construction deformation of rockfill dams[J].Journal of Geotechnical Engineering,1984,110:821-840.
[11]葛哲学,孙志强.神经网络理论与MATLAB R2007实现[M].北京:电子工业出版社,2007.
[12]汪旭,康飞,李俊杰.土石坝地震永久变形参数反演方法研究[J].岩土力学,2014,35(1):279-286.
[13]康飞,李俊杰,许青.堆石坝参数反演的蚁群聚类RBF网络模型[J].岩石力学与工程学报,2009,28(S2):3639-3644.
[14]王柏乐.中国当代土石坝工程[M].北京:中国水利水电出版社,2004.
[2]杨泽艳,周建平,蒋国澄,孙永娟.中国混凝土面板堆石坝的发展[J].水力发电,2011,37(2):18-23.
[3]Behnia D,Ahangari K,Noorzad A,Moeinossadat SR.Predicting crest settlement in concrete face rockfill dams using adaptive neuro-fuzzy inference system and gene expression programming intelligent methods[J].Journal of Zhejiang University SCIENCE A,2013,14(8):589-602.
[4]Zhou W,Hua J,Chang X,Zhou C.Settlement analysis of the Shuibuya concrete-face rockfill dam[J].Computers and Geotechnics,2011,38:269-80.
[5]Seo M-W,Ha IS,Kim Y-S,Olson SM.Behavior of concrete-faced rockfill dams during initial impoundment[J].Journal of Geotechnical and Geoenvironmental Engineering,2009,135:1070-81.
[6]Kim Y-S,Kim B-T.Prediction of relative crest settlement of concrete-faced rockfill dams analyzed using an artificial neural network model[J].Computers and Geotechnics,2008,35:313-322.
[7]周伟,花俊杰,常晓林,杨启贵,马刚.水布垭高面板堆石坝运行期工作性态评价及变形预测[J].岩土工程学报,2011,33(S1):65-70.
[8]詹振彪,罗先启.BP网络与遗传算法在水布垭工程中的应用[J].岩石力学与工程学报,2002,21(7):963-967.
[9]曹克明,汪易森.高混凝土面板堆石坝的设计与施工[J].水力发电,2001(10):49-52.
[10]Clements RP.Post-construction deformation of rockfill dams[J].Journal of Geotechnical Engineering,1984,110:821-840.
[11]葛哲学,孙志强.神经网络理论与MATLAB R2007实现[M].北京:电子工业出版社,2007.
[12]汪旭,康飞,李俊杰.土石坝地震永久变形参数反演方法研究[J].岩土力学,2014,35(1):279-286.
[13]康飞,李俊杰,许青.堆石坝参数反演的蚁群聚类RBF网络模型[J].岩石力学与工程学报,2009,28(S2):3639-3644.
[14]王柏乐.中国当代土石坝工程[M].北京:中国水利水电出版社,2004.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心