摘要
针对河流径流量变化受到众多因素影响,具有随机性和非线性的特征,难以对其精确预测的问题,基于黄河花园口水文站2008—2012年日均流量、日降水量、日均含沙量数据,提出一种结合时间卷积神经网络(TCN)和注意力(Attention)机制的多变量TCN-Attention模型,对花园口水文站日均流量进行预测,并选取LSTM模型和TCN模型进行预测对比实验。结果表明,TCN模型和TCN-Attention模型的预测性能整体优于LSTM模型;Attention机制可以通过调整特征向量权重提升TCN模型的预测性能,与TCN模型相比,TCN-Attention模型的MAE、RMSE、MAPE值分别降低了20.25%、24.90%、24.39%;TCN-Attention模型具有较优的泛化性能,可以提升日均流量预测精度。
In view of the issue that the runoff change was affected by many factors,characterized by randomness and nonlinearity and was dif⁃ficult to predict accurately,based on the data of daily average runoff,daily precipitation and daily average sediment concentration of Hua⁃yuankou Hydrometrical Station on the Yellow River from 2008 to 2012,a multivariable input TCN Attention model combined with time convo⁃lution neural network(TCN)and attention mechanism was proposed to predict the daily runoff of the station,LSTM model and TCN model were selected for prediction and comparison experiments.The results show that the prediction performance of TCN model and TCN Attention model is better than that of LSTM model;The Attention mechanism can improve the prediction performance of TCN model by adjusting the feature vector weight.Compared with TCN model,the MAE,RMSE and MAPE values of TCN Attention model have been decreased by 20.25%,24.90%and 24.39%respectively;TCN Attention model has better generalization performance,which can improve the prediction level of daily runoff.
作者
王军
高梓勋
单春意
WANG Jun;GAO Zixun;SHAN Chunyi(School of Management Engineering,Zhengzhou University,Zhengzhou 450001,China;Institute of Big Data Science,Zhengzhou University of Aeronautics,Zhengzhou 450046,China)
出处
《人民黄河》
CAS
北大核心
2022年第11期20-25,共6页
Yellow River
基金
河南省科技攻关项目(222102210292)
河南省高等学校重点科研项目(20A520041)
河南省重点科技攻关项目(202102210375,212102210518)
河南科技智库调研项目(HNKJZK-2021-61C)。
关键词
日均流量预测
时间卷积神经网络
Attention机制
花园口水文站
average daily flow forecast
temporal convolutional neural network
Attention mechanism
Huayuankou Hydrological Station