基于二次分解和支持向量机的月径流预测方法  

Monthly Runoff Prediction Method Based on Secondary Decomposition and SupportVector Machine

作　　者：甘容 马超鑫 高勇 郭林 侯晓丽 路学永 GAN Rong;MA Chaoxin;GAO Yong;GUO Lin;HOU Xiaoli;LU Xueyong(School of Water Resources and Transportation,Zhengzhou University,Zhengzhou 450001,China;Henan Key Laboratory of Groundwater Pollution Prevention and Rehabilitation,Zhengzhou 450001,China;Henan Provincial Geological Research Institute,Zhengzhou 450001,China;Henan Province Yudong Water Resources Guarantee Center,Kaifeng 475000,China;Canal Head Branch Company of China South-to-North Water Diversion Middle Route Corporation Limited,Nanyang 473000,China)

机构地区：[1]郑州大学水利与交通学院,河南郑州450001 [2]河南省地下水污染防治与修复重点实验室,河南郑州450001 [3]河南省地质研究院,河南郑州450001 [4]河南省豫东水利保障中心,河南开封475000 [5]中国南水北调集团中线有限公司渠首分公司,河南南阳473000

出　　处：《郑州大学学报（工学版）》2024年第6期32-39,共8页Journal of Zhengzhou University（Engineering Science）

基　　金：河南省重点研发与推广专项(232102320026,232102320032);河南省自然资源厅科研项目(202361001);国家自然科学基金资助项目(51509222,51909091)。

摘　　要：针对径流序列的非线性和非平稳性特征,提出了一种基于加权回归的季节趋势分解(STL)和变分模态分解(VMD)组合的二次分解,结合支持向量机(SVM)的月径流预测模型STL-VMD-SVM。该模型利用STL将原始径流序列分解为不同频率的季节项、趋势项和残差项,并通过VMD将残差项分解为IMF s。建立SVM模型预测季节项、趋势项和IMF s,所有IMF s的预测值之和为残差项的预测值,季节项、趋势项和残差项之积为原始径流序列的最终预测值。基于伊洛河流域黑石关站及黄河干流高村站的月径流时间序列进行了实例应用及普适性评价,并与BP神经网络模型和长短期记忆神经网络模型(LSTM)进行对比。结果表明:对于伊洛河黑石关站径流预测,所提模型验证期的NSE、MAPE、RMSE、R分别为0.977,13.705%,0.327,0.991,其预测精度均优于单一模型和一次分解模型,STL-VMD二次分解可以有效提高模型预测精度;在黄河干流高村站径流预测中验证期的NSE、MAPE、RMSE、R分别为0.979,8.509%,3.263,0.989,也达到了很好的预测效果。A monthly runoff prediction model(STL-VMD-SVM)based on a secondary decomposition using loess(STL)and variational mode decomposition(VMD)combined with a support vector machine(SVM)was proposed to address the nonlinear and non-stationary characteristics of runoff sequences.This model utilized STL to decompose the original runoff sequence into seasonal,trend,and residual terms of different frequencies and decomposed the residual term into IMF s through VMD.An SVM model was established to predict seasonal,trend,and IMFs.The sum of the predicted values of all IMF s was the predicted value of the residual term,and the product of seasonal,trend,and residual terms was the final predicted value of the original runoff series.Based on the monthly runoff time series of Heishiguan Station and Gaocun Station on the mainstream of the Yellow River in the Yiluo River Basin,an example application and universality evaluation were conducted,and compared with the BP neural network model and the long shortterm memory neural network model(LSTM).The results showed that for the runoff prediction of Heishiguan Station in the Yiluo River Basin,the NSE,MAPE,RMSE,and R in the validation period of the proposed model were 0.977,13.705%,0.327 and 0.991,respectively,and their prediction accuracy was better than that of the single model and the primary decomposition model.The secondary decomposition of STL-VMD could effectively improve the prediction accuracy of the model.The NSE,MAPE,RMSE,and R during the validation period in the runoff prediction at Gaocun Station on the mainstream of the Yellow River were 0.979,8.509%,3.263,and 0.989,respectively,which also achieved good prediction results.

关 键 词：月径流预测 二次分解 STL VMD SVM 神经网络 

分 类 号：P338[天文地球—水文科学] TV121[水利工程—水文学及水资源]