基于改进GWO-BP神经网络模型的箱涵沉降预测  

Prediction of Box Culvert Settlement Based on Improved GWO-BP Neural Network Model

作　　者：杨阳 赵青[1] 戚蓝[2] 黎启贤 王毓杰 邹爽[1] YANG Yang;ZHAO Qing;QI Lan;LI Qixian;WANG Yujie;ZOU Shuang(School of Civil Engineering,Guizhou University,Guiyang 550025,China;School of Civil Engineering and Architecture,Tianjin University,Tianjin 300350,China;Power China Eco-Environmental Group Co.,Ltd.,Shenzhen 518100,China)

机构地区：[1]贵州大学土木工程学院,贵州贵阳550025 [2]天津大学建筑工程学院,天津300350 [3]中国电建生态环境集团有限公司,广东深圳518100

出　　处：《人民黄河》2021年第10期150-153,共4页Yellow River

基　　金：贵州省自然科学基金资助项目(黔科合J字[2010]2247号);贵州大学引进人才基金资助项目(贵大人基合字[2009]012号);贵州省科技合作计划项目(黔科合LH字[2016]7466号)。

摘　　要：箱涵受外部荷载等多方面影响,会出现一定程度的不均匀沉降,可能会对箱涵结构造成破坏,因此箱涵地基沉降预测十分重要。引入灰狼算法(GWO)对BP神经网络的权值和阈值进行寻优,建立了基于改进的GWO-BP预测模型,对箱涵的沉降值进行预测。将该预测模型应用于南水北调工程天津某标段的箱涵沉降预测,并将预测值与实测值进行对比,相对误差在5%以下。通过与未改进的灰狼算法优化BP神经网络模型、BP模型进行对比,结果表明改进的灰狼算法优化BP神经网络预测模型具有更好的寻优能力与寻优精度,能够有效地对箱涵沉降值进行预测。Due to the influence of external load and other aspects,the box culvert will have a certain degree of uneven settlement,which may cause damage to the box culvert structure.Therefore,the prediction of box culvert foundation settlement is very important.The gray wolf algorithm(GWO)was introduced to optimize the weight and threshold of BP neural network,and the improved GWO-BP prediction model was established to predict the settlement value of box culvert.The prediction model was applied to the settlement prediction of a box culvert in Tianjin section of South-to-North Water Transfer Project,and the relative error was less than 5%.The results show that the Improved Grey Wolf algorithm has better optimization ability and accuracy,and can effectively predict the settlement value of box culvert.

关 键 词：箱涵 沉降预测 灰狼算法 BP神经网络 权值和阈值 

分 类 号：U449.82[建筑科学—桥梁与隧道工程]