基于LSTM算法的大坝坝体渗透压力预测  

Seepage pressure prediction of dam body based on LSTM algorithm

作　　者：姚志武 侯丽娜 文茂华 YAO Zhiwu;HOU Lina;WEN Maohua(Changjiang Survey,Planning,Design and Research Co.,Ltd.,Wuhan 430010,China;Changjiang Space Information Technology Engineering Co.,Ltd.,Wuhan 430010,China)

机构地区：[1]长江勘测规划设计研究有限责任公司,湖北武汉430010 [2]长江空间信息技术工程有限公司(武汉),湖北武汉430010

出　　处：《水利建设与管理》2023年第8期54-59,共6页Water Conservancy Construction and Management

基　　金：国家重点研发计划(2021YFC3200202)。

摘　　要：为更准确地预测大坝坝体渗压变化趋势以确保大坝日常管理安全,利用大量坝体渗压监测数据,构建基于长短期记忆神经网络的大坝坝体渗压预测模型,并与传统BP神经网络方法进行对比。结果表明,LSTM模型能够准确反映大坝坝体渗透系统的不确定性关系,通过对比模型预测值与实际监测值之间的误差,能够初步判定大坝工作状态是否存在安全隐患,可为水库安全运行管理及大坝渗透压力控制提供科学依据,具有一定的工程应用价值。In order to more accurately predict the variation trend of dam seepage pressure and ensure the safety of daily management of the dam,a large amount of dam seepage pressure monitoring data is used to construct a dam seepage pressure prediction model based on long short-term memory neural network and compare it with the traditional Backpropagation(BP)neural network method.The results show that the LSTM model can accurately reflect the uncertainty relationship of the seepage system of the dam body.By comparing the error between the predicted value of the model and the actual monitoring value,it can preliminarily determine whether there is a potential safety hazard in the working state of the dam,which is the basis for the safe operation and management of the reservoir.It provides a scientific basis for the control of seepage pressure and dam seepage pressure,and has good engineering application value.

关 键 词：大坝监测 渗透压力 长短期记忆网络(LSTM) 预测模型 

分 类 号：TV331[水利工程—水工结构工程]