摘要
长期径流预报是保障水利水电工程科学运行管理的重要支撑.随着社会经济的发展,生产实践对长期径流预报精度和预见期长度的更高要求与当前预报的表现出现了矛盾.该研究提出了一种基于深度置信网络(DBN)模型的考虑气候系统指数的长期径流预报框架,并探索了其在预见期1~12个月时的预报表现.结果表明:(1)基于该框架进行预见期为1~12个月的径流预报时,在测试期的纳什效率系数(NSE)值均高于0.50,平均相对误差(MRE)值均低于35%,预报精度较高,且随着预见期的延长,径流预报能力并未呈现下降趋势;(2)在各预见期下,训练期和测试期的预报效果评估指标相差不大,说明采用的基于DBN模型的径流预报框架不存在过拟合的问题,具有优秀的泛化能力;(3)在各预见期下,汛期的径流预报效果在NSE和决定系数R2指标上优于非汛期,在MRE指标上劣于非汛期.
A reliable long-term streamflow forecast is crucial for water resources management and reservoir operations.However,the accuracy of long-term streamflow forecasts are usually unsatisfactory and incapable of meeting stakeholders’needs.In this study,a long-term streamflow forecasting framework based on the deep belief network(DBN)model was proposed and applied to the Tianyi reservoir for monthly streamflow forecasts up to 12 months ahead.The results show that the proposed framework produces reasonable performance with Nash-Sutcliffe efficiency coefficient(NSE)larger than 0.50 and mean relative error(MRE)lower than 35%for all lead months.In addition,the forecasting performance keeps in stable with the extension of the lead time.Moreover,the framework exhibits good generalization ability with no obvious difference between the training and testing periods.Finally,when compared with non-flood season,the forecasting framework exhibits higher NSE and R2 values as well as larger MRE values in the flood season for all lead times.
作者
徐文馨
陈杰
刘建华
陈华
XU Wenxin;CHEN Jie;LIU Jianhua;CHEN Hua(State Key Laboratory of Water Resources Engineering and Management,Wuhan 430072,China;Hubei Provincial Key Lab of Water System Science for Sponge City Construction,Wuhan University,Wuhan 430072,China;TSQ-1 Hydropower Plant of TSQ-1 Hydro Develop Co.,Ltd.,Xingyi 510600,China)
出处
《应用基础与工程科学学报》
EI
CSCD
北大核心
2023年第4期795-810,共16页
Journal of Basic Science and Engineering
基金
国家自然科学基金项目(52079093)
湖北省自然科学基金项目(2020CFA100)。
