张然,柴志勇,张婷,李建柱

• 1:天津大学水利工程仿真与安全国家重点实验室
• 2:中水北方勘测设计研究有限责任公司

摘要(Abstract)：

【目的】准确可靠的洪水预报作为水文预报工作的重中之重，对提高防洪减灾的应急响应效率至关重要。基于数据驱动的机器学习模型无需考虑流域水文物理过程，通过学习历史数据中隐含规律进行建模，在洪水预报领域发展迅速。【方法】将机器学习模型在洪水预报中的应用进行总结，根据结构和复杂度将其分为传统机器学习模型和深度学习模型，包括线性回归、人工神经网络、支持向量机、自适应神经模糊推理系统、小波神经网络和决策树在内的传统机器学习模型，以及包括卷积神经网络和循环神经网络在内的深度学习模型。【结果】结合国内外研究成果，分析了机器学习模型的原理和特点，总结了其在洪水预报中的优势和局限性，并对混合模型的优化方向进行归纳分类，最后对机器学习模型在洪水预报中的未来应用进行展望。【结论】相关成果可为基于机器学习模型的洪水预报研究提供参考。

关键词(KeyWords)： 机器学习;深度学习;数据驱动;洪水预报;水文预报

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(52079086);; 天津市自然科学基金青年项目(20JCQNJC01960)

作者(Author): 张然,柴志勇,张婷,李建柱

DOI: 10.13928/j.cnki.wrahe.2023.11.008

参考文献(References)：

• [1] 王艳艳，陆吉康，陈浩.洪灾损失评估技术的应用[J].水利水电技术，2002(10):30-33.WANG Yanyan,LU Jikang,CHEN Hao.Application of flood damage assessment techniques [J].Water Resources and Hydropower Engineering,2002(10):30-33.
• [2] WANG Z,LAI C,CHEN X,et al.Flood hazard risk assessment model based on random forest[J].Journal of Hydrology,Elsevier,2015,527:1130-1141.
• [3] HAO G,LI J,SONG L,et al.Comparison between the TOPMODEL and the Xin′anjiang model and their application to rainfall runoff simulation in semi-humid regions[J].Environmental Earth Sciences,2018,77(7):279.
• [4] 杨丽洁.数据驱动模型在洪水预报中的应用及其发展趋势[J].电脑知识与技术，2018,14(17):275-277.YANG Lijie.Application of data-driven model in flood forecast and its development trend[J].Computer Knowledge and Technology,2018,14(17):275-277.
• [5] HINTON G E.Reducing the dimensionality of data with neural networks[J].Science,2006,313:504-507.
• [6] 郭丽丽，丁世飞.深度学习研究进展[J].计算机科学，2015,42(5):28-33.GUO Lili,DING Shifei.Research progress on deep learning[J].Computer Science,2015,42(5):28-33.
• [7] KISI ?.Streamflow forecasting using different artificial neural network algorithms[J].Journal of Hydrologic Engineering,2007,12(5):532-539.
• [8] 许剑华，林贞铭.多元线性回归方法在短期洪水预报中的运用[J].水文，1981(6):5-8.XU Jianhua,LIN Zhenming.Application of multiple linear regression method in short-term flood forecasting[J].Journal of China Hydrology,1981(6):5-8.
• [9] 宋润虎.多元线性回归与逐步回归在洪水预报中的应用：以南汀河中长期洪水预报为例[J].人民长江，2012,43(S1):122-125.SONG Runhu.Application of multiple linear regression and stepwise regression in flood forecasting:Nanting River mid-and long-term flood forecasting as an example[J].Yangtze River,2012,43(S1):122-125.
• [10] 张显扬，董增川，王建群，等.雅鲁藏布江尼洋河流域洪水预报方法研究[J].河海大学学报(自然科学版),2005(5):530-533.ZHANG Xianyang,DONG Zengchuan,WANG Jianqun,et al.Method of flood forecasting for Niyang River Basin of Yarlungzangbo River[J].Journal of Hohai University (Natural Sciences),2005(5):530-533.
• [11] MAHABIR C,HICKS F E,ROBICHAUD C,et al.Forecasting breakup water levels at Fort McMurray,Alberta,using multiple linear regression[J].Canadian Journal of Civil Engineering,2006,33(9):1227-1238.
• [12] LATT Z Z,WITTENBERG H.Improving flood forecasting in a developing country:a comparative study of stepwise multiple linear regression and artificial neural network[J].Water Resources Management,2014,28(8):2109-2128.
• [13] JOZAGHI A,SHEN H,GHAZVINIAN M,et al.Multi-model streamflow prediction using conditional bias-penalized multiple linear regression[J].Stochastic Environmental Research and Risk Assessment,2021,35(100):2355-2373.
• [14] VIEIRA A C,SOUZA P,UEYAMA J,et al.Improving flood forecasting through feature selection by a genetic algorithm-experiments based on real data from an Amazon rainforest river[J].Earth Science Informatics,2021,14:37-50.
• [15] 孟宇.灰色关联分析与多元线性回归实时洪水预测模型：以辽河绕阳河流域东白城子站为例[J].中国防汛抗旱，2021,31(10):37-41.MENG Yu.Grey relational analysis and multiple linear regression real-time flood forecasting model:Taking Dongbaichengzi Station in Raoyang River Basin as an example[J].China Flood & Drought Management,2021,31(10):37-41.
• [16] JAIN S K,DAS A,SRIVASTAVA D K.Application of ANN for reservoir inflow prediction and operation[J].Journal of Water Resources Planning & Management,1999,125(5):263-271.
• [17] 马炼，王银堂，张闻胜.神经网络技术在水文系列中长期预报中的应用[J].水利水电技术，2002,33(2):5-7.MA Lian,WANG Yintang,ZHANG Wensheng.Application of neural network technique in medium and long-term hydrologic forecasting [J].Water Resources and Hydropower Engineering,2002,33(2):5-7.
• [18] RANI S,FALGUNI P.Application of artificial neural Network (ANN) for reservoir water level forecasting[J].International Journal of Science and Research,2014,3(7):1077-1082.
• [19] SMITH J,ELI R N.Neural-network models of rainfall-runoff process[J].Journal of Water Resources Planning & Management,1995,121(6):499-508.
• [20] HSU K L,GUPTA H V,SOROOSHIAN S.Artificial neural network modeling of the rainfall:runoff process[J].Water Resources Research,1995,31(10):2517-2530.
• [21] THIRUMALAIAH K,DEO M C.River stage forecasting using artificial neural networks[J].Journal of Hydrologic Engineering,1998,3(1):26-32.
• [22] 胡铁松，袁鹏，丁晶.人工神经网络在水文水资源中的应用[J].水科学进展，1995(1):76-82.HU Tiesong,YUAN Peng,DING Jing.Applications of Artificial Neural Network to hydrology and water resources[J].Advances in Water Science,1995(1):76-82.
• [23] 苑希民，刘树坤，陈浩.基于人工神经网络的多泥沙洪水预报[J].水科学进展，1999(4):393-398.YUAN Ximin,LIU Shukun,CHEN Hao.Application of the Artificial Neural Network method to forecast high sediment percentage flood[J].Advances in Water Science,1999(4):393-398.
• [24] 朱星明，卢长娜，王如云，等.基于人工神经网络的洪水水位预报模型[J].水利学报，2005,36(7):806-811.ZHU Xingming,LU Changna,WANG Ruyun,et al.Artificial Neural Network model for flood water level forecasting[J].Journal of Hydraulic Engineering,2005,36(7):806-811.
• [25] 覃光华，丁晶，刘国东.自适应BP算法及其在河道洪水预报上的应用[J].水科学进展，2002,13(1):37-40.TAN Guanghua,DING Jing,LIU Guodong.River flow prediction using Artificial Neural Networks:Self-adaptive error back-propagation algorithm[J].Advances in Water Science,2002,13(1):37-40.
• [26] 李鸿雁，刘寒冰，苑希民，等.人工神经网络峰值识别理论及其在洪水预报中的应用[J].水利学报，2002,33(6):15-20.LI Hongyan,LIU Hanbing,YUAN Ximin,et al.Peak value recognition theory of Artificial Neural Network and its application to flood forecasting[J].Journal of Hydraulic Engineering,2002,33(6):15-20.
• [27] 金菊良，魏一鸣，杨晓华.基于遗传算法的神经网络及其在洪水灾害承灾体易损性建模中的应用[J].自然灾害学报，1998(2):56-63.JIN Juliang,WEI Yiming,YANG Xiaohua.Genetic algorithm based neural network and its application in modelling the vulnerability of hazard-affected body[J].Journal of Natural Disasters,1998(2):56-63.
• [28] WANG J,SHI P,JIANG P,et al.Application of BP Neural Network algorithm in traditional hydrological model for flood forecasting[J].Water,2017,9(1):48.
• [29] BISWAS R K,JAYAWARDENA A W,HAI P T.Water level forecasting using Artificial Neural Networks[C]//Eleventh International Summer Symposium.Japan:Japan Society of Civil Engineers,2009.
• [30] ZAKARIA M,MALEK M A,ZOLKEPLI M,et al.Application of artificial intelligence algorithms for hourly river level forecast:A case study of Muda River,Malaysia[J].AEJ-Alexandria Engineering Journal,2021,60(4):4015-4028.
• [31] ARANILDO R,LIMA A,ALEX J,et al.Forecasting daily streamflow using online sequential extreme learning machines[J].Journal of Hydrology,2016,537:431-443.
• [32] CHEN L,SUN N,ZHOU C,et al.Flood forecasting based on an improved extreme learning machine model combined with the backtracking search optimization algorithm[J].Water.2018;10(10):1362.
• [33] LI X L,LU H S,AN T Q,et al.Real-time flood forecast using a Support Vector Machine[C]//REN L,WANG W YUAN F.Nanjing:Hydrological Cycle and Water Resources Sustainability in Changing Environments.Int ASSOC Hydrological Sciences,2010.
• [34] 李国英.在2022年全国水利工作会议上的讲话[J].水利发展研究，2022,22(1):1-13.LI Guoying.Speech at the 2022 National Water Conservancy Work Conference[J].Water Resources Development Research,2022,22(1):1-13.
• [35] YU P S,CHEN S T,CHANG I F.Support vector regression for real-time flood stage forecasting[J].Journal of Hydrology,2006,328(3):704-716.
• [36] LIONG S Y,SIVAPRAGASAM C.Flood stage forecasting with support vector machines[J].JAWRA Journal of the American Water Resources Association,2007,38(1):173-186.
• [37] HAN D,CHAN L,ZHU N.Flood forecasting using support vector machines[J].Journal of Hydroinformatics,2007,9(4):267-276.
• [38] CHEN S T,YU P S.Pruning of Support Vector Networks on flood forecasting[J].Journal of Hydrology,2007,347(1-2):67-78.
• [39] 张俊，程春田，申建建，等.基于蚁群算法的支持向量机中长期水文预报模型[J].水力发电学报，2010,29(6):34-40.ZHANG Jun,CHENG Chuntian,SHEN Jianjian,et al.Ant colony optimization-based support vector machine for mid-and-long term hydrological forecasting[J].Journal of Hydroelectric Engineering,2010,29(6):34-40.
• [40] SU J,WANG X,LIANG Y,et al.GA-based support vector machine model for the prediction of monthly reservoir storage[J].Journal of Hydrologic Engineering,2014,19:1430-1437.
• [41] ADNAN R M,YUAN X,KISI O.Stream flow forecasting of poorly gauged mountainous watershed by least square Support Vector Machine,Fuzzy Genetic Algorithm and M5 Model Tree using climatic data from Nearby Station[J].Water Resources Management,2018,32:4469-4486.
• [42] BAFITLHILE T M,LI Z.Applicability of ε-Support Vector Machine and Artificial Neural Network for flood forecasting in Humid,Semi-Humid and Semi-Arid Basins in China[J].Water,2019,11(1):85.
• [43] WU C L,CHAU K W,Li Y S.River stage prediction based on a distributed support vector regression[J].Journal of Hydrology,2008,358(1):96-111.
• [44] 但灵芝，王建群，陈理想，等.v-支持向量机洪水预报模型研究[J].水文，2016,36(2) :7-11.DAN Lingzhi,WANG Jianqun,CHEN Lixiang,et al.Flood forecasting model with ν-Support Vector Machine[J].Journal of China Hydrology,2016,36(2) :7-11.
• [45] NGUYEN D T,CHEN S T.Real-time probabilistic flood forecasting using multiple machine learning methods.Water.2020;12(3):787.
• [46] 周国荣，孔文彬，吴从林，等.基于自适应网络模糊推理方法在河道洪水预报中的应用[J].长江流域资源与环境，2007(5):690-694.ZHOU Guorong,KONG Wenbin,WU Conglin,et al.Application of ANFIS in flood prediction in main stream of the Yangtze River[J].Resources and Environment in the Yangtze Basin,2007(5):690-694.
• [47] CHEN S H,LIN Y H,CHANG L C,et al.The strategy of building a flood forecast model by neuro-fuzzy network[J].Hydrological Process,2006,20:1525-1540.
• [48] SHU C,OUARDA T.Regional flood frequency analysis at ungauged sites using the adaptive neuro-fuzzy inference system[J].Journal of Hydrology,2008,349(1-2):31-43.
• [49] LOHANI A K,KUMAR R,SINGH R D.Hydrological time series modeling:A comparison between adaptive neuro-fuzzy,neural network and autoregressive techniques[J].Journal of Hydrology,2012,442-443:23-35.
• [50] 杜富慧，王晓丽，郝振纯.基于ANFIS的黄河灵口流域水文预报模型[J].人民黄河，2012,34(6):41-42.DU Fuhui,WANG Xiaoli,HAO Zhenchun.Flood forecasting model based on ANFIS in Lingkou Catchment of the Yellow River[J].Yellow River,2012,34(6):41-42.
• [51] 路剑飞，陈子燊，王扬圣.基于MI-ANFIS的中长期水文预报方案优选研究[J].水力发电学报，2013,32(2):48-53.LU Jianfei,CHEN Ziyan,WANG Yangsheng.Optimization selection of mid and long-term hydrologic forecast schemes based on MI-ANFIS[J].Journal of Hydroelectric Engineering,2013,32(2):48-53.
• [52] 唐英敏，闫强.小波-ANFIS模型在年最大洪峰预测中的应用[J].人民黄河，2015,37(10):42-45.TANG Yingmin,YAN Qiang.Application and research of annual maximum peak discharge forecast based on Wavelet-ANFIS[J].Yellow River,2015,37(10):42-45.
• [53] BUI D T,PANAHI M,SHAHABI H,et al.Novel hybrid evolutionary algorithms for spatial prediction of floods[J].Entific Reports,2018,8(1):15364.
• [54] KIM B,CHOI S Y,HAN K Y.Integrated real-time flood forecasting and inundation analysis in small-medium streams[J].Water,2019,11(5):919.
• [55] 刘国华，钱镜林，王建江.小波软阈值技术和人工神经网络在洪水预报中的研究[J].水力发电学报，2004(4):5-10.LIU Guohua,QIAN Jinglin,WANG Jianjiang.Study of flood forecast based on wavelet soft-threshold technology and ANN[J].Journal of Hydroelectric Engineering,2004(4):5-10.
• [56] 李致家，周轶，李志龙，等.小波变换与BP神经网络耦合的洪水预报方法[J].水力发电学报，2009,28(2):20-24.LI Zhijia,ZHOU Yi,LI Zhilong,et al,Flood forecasting method of coupling wavelet transform with BP neural network[J].Journal of Hydroelectric Engineering,2009,28(2):20-24.
• [57] SEO Y,KIM S,KISI O,et al.Daily water level forecasting using wavelet decomposition and artificial intelligence techniques[J].Journal of Hydrology,2015,520:224-243.
• [58] SAHAY R R,SRIVASTAVA A.Predicting monsoon floods in rivers embedding wavelet transform,Genetic Algorithm and Neural Network[J].Water Resources Management,2014,28(2):301-317.
• [59] TIWARI M K,CHATTERJEE C.Development of an accurate and reliable hourly flood forecasting model using wavelet-bootstrap-ANN (WBANN) hybrid approach[J].Journal of Hydrology,2010,394(3-4):458-470.
• [60] KASIVISWANATHAN K S,HE J,SUDHEER K P,et al.Potential application of wavelet neural network ensemble to forecast streamflow for flood management[J].Journal of Hydrology,2016,536:161-173.
• [61] 丁胜祥，董增川，张莉.基于决策树算法的洪水预报模型[J].水力发电，2011,37(7):8-11.DING Shengxiang,DONG Zengchuan,ZHANG Li.Flood forecasting model based on decision tree algorithm[J].Water Power,2011,37(7):8-11.
• [62] HAN D,CLUCKIE I D,KARBASSIOUN D,et al.River flow modelling using fuzzy decision trees[J].Water Resources Management,2002,16(6):431-445.
• [63] TEHRANY M S,PRADHAN B,JEBUR M N.Spatial prediction of flood susceptible areas using rule based decision tree (DT) and a novel ensemble bivariate and multivariate statistical models in GIS[J].Journal of Hydrology,2013,504:69-79.
• [64] KHOSRAVI K,PHAM B T,CHAPI K,et al.A comparative assessment of decision trees algorithms for flash flood susceptibility modeling at Haraz watershed,northern Iran[J].Science of The Total Environment,2018,627:744-755.
• [65] WEI C C.Discretized and continuous target fields for the reservoir release rules during floods[J].Water Resources Management,2012,26:3457-3477.
• [66] 姚超宇，钟平安，徐斌，等.基于GBDT的实时洪水预报误差校正方法[J].水电能源科学，2019,37(8):38-42.YAO Chaoyu,ZHONG Pingan,XU Bin,et al.Real-time correction method of flood forecasting error based on GBDT[J].Water Resources and Power,2019,37(8):38-42.
• [67] LI B,YANG G,WAN R,et al.Comparison of random forests and other statistical methods for the prediction of lake water level:a case study of the Poyang Lake in China[J].Hydrology Research,2016,47(S1):69-83.
• [68] 张珂，牛杰帆，李曦，等.洪水预报智能模型在中国半干旱半湿润区的应用对比[J].水资源保护，2021,37(1):28-35.ZHANG Ke,NIU Jiefan,LI Xi,et al.Comparison of artificial intelligence flood forecasting models in China′s semi-arid and semi-humid regions[J].Water Resources Protection,2021,37(1):28-35.
• [69] 赖成光，陈晓宏，赵仕威，等.基于随机森林的洪灾风险评价模型及其应用[J].水利学报，2015,46(1):58-66.LAI Chengguang,CHEN Xiaohong,ZHAO Shiwei,et al.A flood risk assessment model based on Random Forest and its application[J].Journal of Hydraulic Engineering,2015,46(1):58-66.
• [70] GANG Z,BO P A,ZX A,et al.Mapping flood susceptibility in mountainous areas on a national scale in China[J].Science of the Total Environment,2018,615:1133-1142.
• [71] CHEN W,LI Y,XUE W,et al.Modeling flood susceptibility using data-driven approaches of nave Bayes tree,alternating decision tree,and random forest methods[J].Science of the total Environment,2019,701,134979.
• [72] CHEN C,HUI Q,XIE W,et al.Convolutional Neural Networks for forecasting flood process in Internet-of-Things enabled smart city[J].Computer Networks,2020,186:107744.
• [73] WANG J H,LIN G F,CHANG M J,et al.Real-time water-level forecasting using dilated causal Convolutional Neural Networks[J].Water Resources Management,2019,33:3759-3780.
• [74] 姬战生，章国稳，张振林.基于卷积神经网络的东苕溪瓶窑水文站水位预报[J].水电能源科学，2021,39(8):46-49.JI Zhansheng,ZHANG Guowen,ZHANG Zhenlin.Water level forecast of Pingyao Hydrological Station in Dongtiaoxi based on Convolutional Neural Network[J].Water Resources and Power,2021,39(8):46-49.
• [75] KABIR S R,PATIDAR S,XIA X.A deep convolutional neural network model for rapid prediction of fluvial flood inundation[J].Journal of Hydrology,2020,590:125481.
• [76] HOSSEINY H.A deep learning model for predicting river flood depth and extent[J].Environmental Modelling and Software,2021,145:105186.
• [77] ROLAND L,JULIAN B,DAVID G J,et al.U-FLOOD-Topographic deep learning for predicting urban pluvial flood water depth[J].Journal of Hydrology,2021,603:126898.
• [78] CHIANG Y M.Merging multiple precipitation sources for flash flood forecasting[J].Journal of Hydrology,2007,340 (3-4):183-196.
• [79] 李原园，张宜清，杨晓茹.完善流域防洪工程体系提升洪涝灾害防御能力[J].水利发展研究，2021,21(9):28-29.LI Yuanyuan,ZHANG Yiqing,YANG Xiaoru.Improving the Basin Flood Control Engineering System and Enhancing the Flood Disaster Prevention Ability[J].Water Resources Development Research,2021,21(9):28-29.
• [80] LE X H,HO H V,LEE G,et al.Application of Long Short-Term Memory (LSTM) Neural Network for flood forecasting[J].Water,2019,11(7):1387.
• [81] 舒全英，郭磊，孙甜，等.基于数据驱动的碗窑水库洪水预报[J].水电能源科学，2021,39(11):115-118.SHU Quanying,GUO Lei,SUN Tian,et al.Wanyao Reservoir flood forecasting based on data driven[J].Water Resources and Power,2021,39(11):115-118.
• [82] LIU Y,WANG H,FENG W,et al.Short term real-time rolling forecast of urban river water levels based on LSTM:A case study in Fuzhou City,China[J].International Journal of Environmental Research and Public Health,2021,18(17):9287.
• [83] WU Y,LIU Z,XU W,et al.Context-aware attention LSTM network for flood prediction[C]//IEEE:2018 24th International Conference on Pattern Recognition (ICPR).Beijing:IEEE,2018.
• [84] DING Y K,ZHU Y L,Feng J.Interpretable spatio-temporal attention LSTM model for flood forecasting[J].Neurocomputing,2020,403:348-359.
• [85] LYU N,LIANG X,CHEN C.A long short-term memory cyclic model with mutual information for hydrology forecasting:A case study in the Xixian Basin[J].Advances in Water Resources,2020,141:103622.
• [86] 张甲甲，万定生.基于混合GA优化LSTM的中小流域流量预测研究[J].计算机仿真，2022,39(2):283-287.ZHANG Jiajia,WAN Dingsheng.Improved hybrid genetic algorithm and its application in runoff prediction[J].Computer Simulation,2022,39(2):283-287.
• [87] KAO I F,ZHOU Y,CHANG L C,et al.Exploring a Long Short-Term Memory based Encoder-Decoder framework for multi-step-ahead flood forecasting[J].Journal of Hydrology,2021,583:124631.
• [88] CUI Z,ZHOU Y,GUO S.A novel hybrid XAJ-LSTM model for multi-step-ahead flood forecasting[J].Hydrology Research,2021,52 (6):1436-1454.
• [89] MOISHIN M,DEO R C,PRASAD R.Designing deep-based learning flood forecast model with ConvLSTM hybrid algorithm[J].IEEE Access,2021,9:50982-50993.
• [90] HU R,FANG F,PAIN C C.Rapid spatio-temporal flood prediction and uncertainty quantification using a deep learning method[J].Journal of Hydrology,2019,575:911-920.
• [91] MIAU S,HUNG W H.River Flooding forecasting and anomaly detection based on deep learning[J].IEEE Access,2020,8:198384-198402.
• [92] GAO S,HUANG Y,ZHANG S,et al.Short-term runoff prediction with GRU and LSTM networks without requiring time step optimization during sample generation[J].Journal of Hydrology,2020,589:125188.
• [93] 郭燕，赖锡军.基于循环神经网络的洞庭湖水位预测研究[J].长江流域资源与环境，2021,30(3):689-698.GUO Yan,LAI Xijun.Research on water level prediction of Dongting Lake based on recurrent neural network[J].Resources and Environment in the Yangtze Basin,2021,30(3):689-698.