摘要
水文预报是水资源优化配置的重要前提,而传统预报方法普遍存在预测精度低的问题,为提高水文预报的准确性,提出了一种混合数据驱动模型用于月径流预测,即奇异谱分析-灰狼优化-支持向量回归(SSA-GWO-SVR)模型。该模型通过SSA对径流数据进行去噪处理来提高径流序列的平稳性和可预测性,采用GWO对SVR模型的参数进行联合选优,从而增强模型的泛化能力。通过黑河正义峡的月径流预测进行模型验证,以平均绝对误差(MAE)、均方根误差(RMSE)、相关系数(R)和纳什效率系数(NSEC)为模型评价标准。实验结果表明该模型的预测精度明显高于自回归积分滑动平均模型(ARIMA)、持续性模型(PM)、交叉验证-SVR(CV-SVR)和GWOSVR模型,并且它能很好地预测径流峰值,说明该模型是一种可靠的径流预测模型,能够更深入地捕获水文径流的内在特性,为基于数据驱动模型的水文预报提供了一种新方法。
Hydrologic prediction is an important prerequisite for optimal allocation of water resources,but the traditional forecasting methods generally have the problem of low forecasting accuracy.To improve the accuracy of hydrologic prediction,a hybrid data-driven model is proposed for monthly runoff forecasting,namely,Singular Spectrum Analysis-Grey Wolf Optimizer-Support Vector Regression(SSA-GWO-SVR)model.The proposed model uses SSA to denoise the runoff data to improve the stability and predictability of runoff series,and uses GWO to optimize the parameters of SVR model to enhance the generalization ability of the model.This model is validated by monthly runoff prediction of Zhengyixia in the Heihe River Basin,and the Mean Absolute Error(MAE),Root Mean Square Error(RMSE),correlation coefficient(R)and Nash-Sutcliffe Efficiency Coefficien(NSEC)are used as evaluation criteria.The experimental results show that the prediction accuracy of the proposed model is significantly higher than those of Autoregressive Integrated Moving Average model(ARIMA),Persistent Model(PM),Cross Validation(CV)-SVR and GWO-SVR models,and the can predict the runoff peak well,which indicates that the model is a reliable runoff forecasting model,can capture the intrinsic characteristics of hydrologic runoff more deeply,and provides a new method for hydrologic prediction based on data-driven model.
作者
王丽丽
李新
冉有华
郭彦龙
Wang Lili;Li Xin;Ran Youhua;Guo Yanlong(Key Laboratory of Remote Sensing of Gansu Province,Northwest Institute of Eco-Environment and Resources,Chinese Academy of Sciences,Lanzhou 730000,China;College of Physics and Electrical Engineering,Northwest Normal University,Lanzhou 730070,China;University of Chinese Academy of Sciences,Beijing 100049,China;Institute of Tibetan Plateau Research,Chinese Academy of Sciences,Beijing 100101,China;CAS Center for Excellence in Tibetan Plateau Earth Sciences,Chinese Academy of Sciences,Beijing 100101,China)
出处
《遥感技术与应用》
CSCD
北大核心
2020年第2期355-364,共10页
Remote Sensing Technology and Application
基金
中国科学院战略性先导科技专项(XDA19070104、XDA20100104)
中国科学院信息化项目(XXH13505-06)
国家自然科学基金项目(41630856)
西北师范大学青年教师科研能力提升计划项目(NWNU-LKQN2019-18)资助
关键词
径流预测
数据驱动
奇异谱分析
灰狼优化
支持向量回归
Runoff prediction
Data-driven
Singular spectrum analysis
Grey wolf optimizer
Support vector regression
