首页|基于混沌云量子蝙蝠CNN-GRU大坝变形智能预报方法研究

基于混沌云量子蝙蝠CNN-GRU大坝变形智能预报方法研究

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
针对大坝变形影响因素复杂、精准预报难度较大问题,为了提高在大坝安全管理过程中大坝变形的预报精度,本文从大坝变形非线性动力系统时间序列的强非线性出发,引入深度卷积神经网络,对大坝变形及其空间影响特性进行挖掘,引入门控循环单元,对大坝变形的时域特性进行挖掘,构建应用于大坝变形预报的深度卷积神经网络-门控循环单元大坝变形组合深度学习网络;同时,为了获取深度卷积神经网络-门控循环单元组合网络的最佳超参,引入了混沌云量子蝙蝠算法,建立了基于混沌云量子蝙蝠算法算法的深度卷积神经网络-门控循环单元组合网络超参优选方法;最后,提出了深度卷积神经网络-门控循环单元-混沌云量子蝙蝠算法大坝变形组合深度学习智能预报方法.基于实测数据开展预报研究,对比结果表明:与对比模型相比,提出的深度卷积神经网络-门控循环单元-混沌云量子蝙蝠算法预报方法取得了更精确的预报结果,混沌云量子蝙蝠算法算法用于超参优选获得了更佳的超参组合.
Research on the intelligent prediction method of dam deformation based on chaotic cloud quantum bat CNN-GRU
The factors influencing dam deformation are complex,and an accurate prediction is difficult.This study aims to improve the prediction accuracy of dam deformation in dam safety management.Commencing from the strong nonlinearity of the time series of the nonlinear dynamics system of dam deformation,this paper introduces deep convolutional neural networks(CNN)to explore the characteristics of dam deformation and its spatial influ-ence as well as a gated recurrent unit(GRU)to explore the time-domain characteristics of dam deformation to con-struct a CNN-GRU combined deep learning network of dam deformation for its timely prediction.In addition,to ob-tain the optimal superparameters of the CNN-GRU combined network,the chaotic cloud quantum bat algorithm(CCQBA)is introduced,and a superparameter optimization method of the CNN-GRU combined network based on the CCQBA is established.Finally,a combined deep learning intelligent prediction method of CNN-GRU-CCQBA dam deformation is proposed,and the prediction research is conducted based on the measured data.The compari-son results show that in contrast to the comparison model,the proposed CNN-GRU-CCQBA prediction method yields more accurate prediction results,and a superior hyperparameter combination is obtained when the CCQBA is used for hyperparameter optimization.

dam deformation predictionconvolutional neural networkgated recurrent unitbat algorithmquan-tum mechanicschaos theorysimulation and prediction of nonlinear dynamics systemdeep learning

陈以浩、李明伟、安小刚、王宇田、徐瑞喆

展开 >

交通运输部水运科学研究院,北京 100088

哈尔滨工程大学 船舶工程学院,黑龙江 哈尔滨 150001

大坝变形预测 卷积神经网络 门控循环单元 蝙蝠算法 量子力学 混沌理论 非线性动力系统模拟与预测 深度学习

国家自然科学基金国家重点基础研究发展计划(973计划)黑龙江省优秀青年基金

515090562019TFB1504403YQ2021E015

2024

10.11990/jheu.202202004

哈尔滨工程大学学报
哈尔滨工程大学

哈尔滨工程大学学报

CSTPCD北大核心
影响因子：0.655
ISSN：1006-7043
年,卷(期)：2024.45(1)
  • 1
  • 14