基于MK-SVM和时序特征分析的月径流预报模型  

Monthly runoff prediction model based on MK-SVM and time series feature analysis

作　　者：雷庆文 闫磊 巫晨煜 罗云 谢笑添 LEI Qingwen;YAN Lei;WU Chenyu;LUO Yun;XIE Xiaotian(College of Water Conservancy and Hydropower,Hebei University of Engineering,Handan 056038,China;Hebei Key Laboratory of Intelligent Water Conservancy,Handan 056038,China;Yunhe(Henan)Information Technology Co.,Ltd.,Zhengzhou 450003,China;Institute of International Rivers and Eco-security,Yunnan University,Kunming 650091,China)

机构地区：[1]河北工程大学水利水电学院,河北邯郸056038 [2]河北省智慧水利重点实验室,河北邯郸056038 [3]云河(河南)信息科技有限公司,河南郑州450003 [4]云南大学国际河流与生态安全研究院,云南昆明650091

出　　处：《水资源保护》2024年第6期148-154,共7页Water Resources Protection

基　　金：国家自然科学基金项目(51909053);水利部京津冀水安全保障重点实验室开放研究基金项目(IWHR-KLWS-202305)。

摘　　要：针对传统径流预报方法预报因子不确定性和预报模型复杂性问题,基于月径流时序特征重要性分析选择预报因子,采用混合核函数支持向量机(MK-SVM)模型捕捉径流时序间的非线性关系,提出动态透镜成像反向学习和Lévy飞行等多策略融合的改进灰狼优化算法(IGWO),并构建了径流预报的IGWO-MK-SVM模型。黑河流域莺落峡水文站月径流预报结果表明:IGWO-MK-SVM模型月径流预报结果的纳什效率系数、均方根误差、Kling-Gupta效率系数分别为0.8942、16.9099 m^(3)/s和0.8639;与传统SVM模型相比,IGWO-MK-SVM模型在径流预报中的自适应性有所提升,相较于长短期记忆网络模型和季节性差分自回归移动平均模型,IGWO-MK-SVM模型能更好地预报月径流的真实变化过程。To address the problem of uncertainty of prediction factors and model complexity of traditional runoff prediction methods,prediction factors were selected based on feature importance analysis of monthly runoff time series,and the nonlinear relationship between runoff time series was captured by the mixed kernel function-support vector machine(MK-SVM)model.An improved grey wolf optimizer(IGWO)that integrated multiple strategies,such as dynamic lens imaging reverse learning and Lévy flying strategies,was proposed to enhance the stability of the global parameter optimization of the MK-SVM model,and an IGWO-MK-SVM model for runoff prediction was constructed.The results of monthly runoff prediction at Yingluoxia Hydrological Station in the Heihe River Basin show that the Nash-Sutcliffe efficiency coefficient,root mean squared error,and Kling-Gupta efficiency coefficient of prediction results of the IGWO-MK-SVM model were 0.8942,16.9099 m^(3)/s,and 0.8639,respectively.Compared with the traditional SVM model,the IGWO-MK-SVM model has high adaptability in runoff prediction,and compared with the long short-term memory network model and the seasonal autoregressive integrated moving average model,the IGWO-MK-SVM model can better predict the real change process of monthly runoff.

关 键 词：径流预报 随机森林 径流预报因子 混合核函数支持向量机 改进灰狼优化算法 黑河流域 

分 类 号：P338[天文地球—水文科学]