基于卷积神经网络的大坝位移预测模型对比研究  

Comparison of dam displacement prediction models based on convolutional neural network

作　　者：沈忱 周娟 SHEN Chen;ZHOU Juan(Anhui Shuian Construction Group Co.,Ltd,Hefei 230026,China;Anhui Technical College of Water Resources and Hydroelectric Power,Hefei 231603,China)

机构地区：[1]安徽水安建设集团股份有限公司,安徽合肥230026 [2]安徽水利水电职业技术学院,安徽合肥231603

出　　处：《安徽水利水电职业技术学院学报》2023年第1期5-8,69,共5页Journal of Anhui Technical College of Water Resources and Hydroelectric Power

基　　金：安徽省高校自然科学研究重点项目(KJ2020A1046)。

摘　　要：针对当前大坝位移预测研究中对大坝变形影响因子考虑欠缺的问题,将卷积神经网络与典型机器学习算法相融合对大坝位移进行预测。结合某大坝2018年3月1日至2022年5月1日的监测数据进行研究,并将大坝位移预测值与经典随机森林预测结果、支持向量机预测结果进行比较,结果表明融合卷积神经网络后的模型比单一模型具有更高的预测精度和更小的预测误差,模型预测性能更好。In view of the lack of consideration of the influence factors of dam deformation in the current dam displacement prediction research,the convolutional neural network and the typical machine learn-ing algorithm are combined to predict the dam displacement.Combined with the monitoring data of a dam from March 1,2018 to May 1,2022,the research is carried out,and the dam displacement pre-diction value is compared with the classic random forest prediction results and support vector machine prediction results.The results show that the model after the fusion of convolutional neural network has higher prediction accuracy and smaller prediction error than the single model,and the model pre-diction performance is better.

关 键 词：大坝变形 位移预测 卷积神经网络 

分 类 号：TV698.1[水利工程—水利水电工程]