Simulation of Daily Runoff at the Yellow River Source Region Using a Combined LSTM-KNN Model LI Yonghua1,2, DAI Qingcuo3, MA Yufang3, LIU Wei3, ZHANG Jing3
This study develops a hybrid model integrating Long Short-Term Memory (LSTM) and K-Nearest Neighbors (KNN) algorithms to predict daily runoff at the Jimai Station, Jungong Station, and Tangnaihai Station in the Yellow River source region. Dynamic attributes of the watershed were constructed using meteorological variables (e.g., temperature and precipitation), while static attributes were derived from historical hydro-meteorological and geographic data. Feature selection was performed using the LSTM model, and the optimized TOPO_CLIM_SOIL_LSTM model was applied for daily runoff prediction, followed by real-time correction via the KNN algorithm. Results indicate that the TOPO_CLIM_SOIL_LSTM model effectively captures rainfall-runoff relationships and stabilizes low-flow predictions. After KNN correction, the accuracy of next-day runoff forecasts exceeds 93% at all stations, with the Nash-Sutcliffe Efficiency (NSE) increasing by 18.07%, 6.45%, and 12.5% for Jimai Station, Jungong Station, and Tangnaihai Station, respectively, demonstrating significant improvement in prediction precision.
XIAOFengjin, XUYuqing, HUANGDapeng,et al.Impact of climate change on ecological security of the Yellow River Basin and its adaptation countermeasures[J].Yellow River,2021,43(1):10-14, 52.(in Chinese)
HANHuibang, ZHANGBoyue, MAShoucun,et al.A study on the distribution of raindrop size in the upper Yellow River[J].Desert and Oasis Meteorology,2019,13(6):119-125.(in Chinese)
JIZhejun, WANGLina, LIGuojun,et al.Analysis on changes of cloud coverage in the Maqu Area of the upper reaches of Yellow River in the recent 40 years[J].Desert and Oasis Meteorology,2014,8(5):29-33.(in Chinese)
LIGuojun, LIXiaoyuan, WANGZhenguo,et al.Climate change and its impact on water resource in water replenishment region in upper stream of Yellow River[J].Arid Meteorolgy,2007,25(2):67-70, 89.(in Chinese)
[9]
LIUYongqi, YELei, QINHui,et al.Monthly streamflow forecasting based on hidden Markov model and Gaussian mixture regression[J].J Hydrol,2018,561:146-159.
[10]
左岗岗.基于机器学习的径流预测方法及适应性预测机制研究[D].西安:西安理工大学,2021.
[11]
ZUOGanggang.Study on mechanism and methods of adaptive streamflow forecasting based on machine learning[D].Xi'an:Xi'an University of Technology,2021.(in Chinese)
[12]
LUOXiangang, YUANXiaohui, ZHUShuang,et al.A hybrid support vector regression framework for streamflow forecast[J].J Hydrol,2019,568:184-193.
[13]
MoradkhaniH, HsuK L, GuptaH V,et al.Improved streamflow forecasting using self-organizing radial basis function artificial neural networks[J].J Hydrol,2004,295(1/4):246-262.
[14]
GranataF, Di NunnoF, De MarinisG.Stacked machine learning algorithms and bidirectional long short-term memory networks for multi-step ahead streamflow forecasting:a comparative study[J].J Hydrol,2022,613:128431.
XIEXiaoyan, LIChunhong, WANGJianping,et al.Research on combined forecast model of medium and long-term hydrological forecast based on the combination of ANN and multivariate threshold regression[J].Hydropower Plant Automation,2013(4):45-47, 60.(in Chinese)
ZHANGChi, ZHOUHuicheng, WANGBende,et al.Combined forecast method for classified forecast of river flood propagation[J].Journal of Harbin Institute of Technology,2008,40(8):1307-1310.(in Chinese)
LIBin, WEIYinwu, QIBing,et al.DDoS attack detection method based on EEMD-LSTM for demand response terminal[J].Electric Power Construction,2022,43(4):81-90.(in Chinese)
[25]
HeZ H, ParajkaJ, TianF Q,et al.Estimating degree-day factors from MODIS for snowmelt runoff modeling[J].Hydrol Earth Syst Sci,2014,18(12):4773-4789.
XUDongmei, LIAOAndong, WANGWenchuan.Monthly runoff prediction based on VMD-EEMD-CNN-LSTM mixed model[J].Water Resources Planning and Design,2023(7):57-63.(in Chinese)
ZHANGShaoqing, CHENYipu, WANGShihui,et al.A runoff prediction method based on multi-scale deep learning of differential sequence[J].Water Power,2024,50(4):19-25.(in Chinese)
[30]
田杨.基于长短期记忆神经网络的舞水河径流量预测研究[D].长沙:湖南农业大学,2021:51.
[31]
TIANYang.Wushui River runoff series prediction research based on long-short term memory neural network[D].Changsha:Hunan Agricultural University,2021:51.(in Chinese)
ZHANGMengfan, DINGBingbing, JIAGuodong,et al.Comparative prediction of runoff in the Beiluo River,Shaanxi Province of northwestern China based on TCN-BiLSTM and LSTM models[J].Journal of Beijing Forestry University,2024,46(4):141-148.(in Chinese)
XIONGYi, ZHOUJianzhong, SUNNa,et al.Monthly runoff prediction based on self-adaptive variational mode decomposition and long short-term memory network[J].Journal of Hydraulic Engineering,2023,54(2):172-183, 198.(in Chinese)