大连理工大学主页平台管理系统 dingwei--Home-- Flash Flood Forecasting Based on Long Short-Term Memory Networks

Paper Publications

Flash Flood Forecasting Based on Long Short-Term Memory Networks

Hits:

Indexed by:Journal Papers

Date of Publication:2020-01-01

Journal:WATER

Included Journals:EI、SCIE

Volume:12

Issue:1

Key Words:flash flood forecasting; long short-term memory; recurrent neural networks; machine learning

Abstract:Flash floods occur frequently and distribute widely in mountainous areas because of complex geographic and geomorphic conditions and various climate types. Effective flash flood forecasting with useful lead times remains a challenge due to its high burstiness and short response time. Recently, machine learning has led to substantial changes across many areas of study. In hydrology, the advent of novel machine learning methods has started to encourage novel applications or substantially improve old ones. This study aims to establish a discharge forecasting model based on Long Short-Term Memory (LSTM) networks for flash flood forecasting in mountainous catchments. The proposed LSTM flood forecasting (LSTM-FF) model is composed of T multivariate single-step LSTM networks and takes spatial and temporal dynamics information of observed and forecast rainfall and early discharge as inputs. The case study in Anhe revealed that the proposed models can effectively predict flash floods, especially the qualified rates (the ratio of the number of qualified events to the total number of flood events) of large flood events are above 94.7% at 1-5 h lead time and range from 84.2% to 89.5% at 6-10 h lead-time. For the large flood simulation, the small flood events can help the LSTM-FF model to explore a better rainfall-runoff relationship. The impact analysis of weights in the LSTM network structures shows that the discharge input plays a more obvious role in the 1-h LSTM network and the effect decreases with the lead-time. Meanwhile, in the adjacent lead-time, the LSTM networks explored a similar relationship between input and output. The study provides a new approach for flash flood forecasting and the highly accurate forecast contributes to prepare for and mitigate disasters.

Pre One:Reference Point Based Multi-Objective Optimization of Reservoir Operation: a Comparison of Three Algorithms

Next One:Tradeoff Analysis Index for Many-Objective Reservoir Optimization

Profile

丁伟，副教授，博导，长期从事流域水资源管理研究，主持国家自然科学基金项目青年基金1项、面上项目1项，十三五国家重点研发计划子课题1项、十四五国家重点研发计划子课题1项，作为技术骨干参与了国家自然科学基金重点项目、国际合作重点项目等多项课题。

发表SCI/EI论文30余篇，其中水文水资源领域顶级期刊《Water Resources Research》4篇（均为1作/通讯）、ASCE百年旗舰期刊《Journal of Water Resources Planning and Management》 4篇（3篇为1作/通讯）；授权国家发明专利4项、国际发明专利1项，且有1项实现百万成果转化；出版专著2部；获2019年辽宁省科技进步一等奖、2016年教育部科技进步一等奖、2021年大坝工程学会科技进步奖一等奖1项。

代表性论文

[1] Wei Ding, Bing Yu, Yong Peng^*, Guang Han, Lin Zhang, The marginal utility principles for a two-stage hydropower operation problem. Journal of Water Resources Planning and Management, 2022，DOI:10.1061/(ASCE)WR.1943-5452.0001556

[2] Xingsheng Shu, Wei Ding^*, Yong Peng^*, Ziru Wang, Jian Wu, Min Li. Monthly Streamflow Forecasting Using Convolutional Neural Network. Water Resources Management, 2021, 35(15): 5089-5104, DOI: 10.1007/s11269-021-02961-w

[3] Yu Li, Wei Ding^*, Xiaoxian Chen, Ximing Cai, Chi Zhang. An analytical framework for reservoir operation with combined natural inflow and controlled inflow, Water Resources Research,2020,56(8),e2019WR025347, Doi: 10.1029/2019WR025347

[4]Wei Ding, Haixing Liu^*, Yu Li, Hua Shang, Chi Zhang, Guangtao Fu. Unraveling the effects of long-distance water transfer for ecological recharge, Journal of Water Resources Planning and Management,2020,146(9):02520004, DOI: 10.1061/(ASCE)WR.1943-5452.0001272

[5] Bingyao Zhang, Wei Ding^*, Bo Xu, Longfan Wang, Yu Li, Chi Zhang. Spatial characteristics of total phosphorus loads from different sources in the Lancang River Basin, Science of The Total Environment,2020,722,137863, DOI: 10.1016/j.scitotenv.2020.137863

[6] Chi Zhang, Wei Ding^*, Yu Li, Fanlin Meng, Guangtao Fu. Cost-Benefit framework for optimal design of water transfer systems, Journal of Water Resources Planning and Management,2019,145(5):04019007, DOI:10.1061/(ASCE)WR.1943-5452.0001059

[7] Qiang Wang, Wei Ding^*, Yan Wang. Optimization of multi-reservoir operating rules for a water supply system, Water Resources Management,2018,32(14):4543-4559, DOI: 10.1007/s11269-018-2063-9

[8] Wei Ding, Chi Zhang ^*, Ximing Cai, Yu Li, Huicheng Zhou, Multiobjective hedging rules for flood water conservation, Water Resources Research, 2017,53(3):1963-1981, DOI: 10.1002/2016WR019452

[9] Chi Zhang, Wei Ding^*, Yu Li, Yin Tang, Dingbao Wang. Catchments' hedging strategy on evapotranspiration for climatic variability, Water Resources Research,2016,52(11):9036-9045, DOI：10.1002/2016WR019384

[10] Wei Ding, Chi Zhang^*, Yong Peng, Ruijie Zeng, Huicheng Zhou, Ximing Cai. An analytical framework for flood water conservation considering forecast uncertainty and acceptable risk, Water Resources Research,2015,51(6):4702-4726, DOI: 10.1002/2015WR017127

科研项目

[1] 国家重点研发计划课题，特大干旱条件下旱限（警）水位确定与抗旱应急智慧调度研究子课题：特大干旱下水利工程群应急调度技术研究，项目负责人

[2] 国家自然科学基金面上项目，基于知识学习的水库防洪智能调度研究，项目负责人

[3] 国家自然科学基金青年基金，水电站汛期防洪与发电协调控制理论与方法研究，项目负责人

[4] 国家重点研发计划，长江上游梯级水库群多目标联合调度技术子课题：面向两湖和长江口地区供水需求的水库群供水调度方案研究，项目负责人

[5] 水利前期项目，全国江河湖库旱限水位（流量）确定试点，技术负责人

[6] 吉林省农村基层项目，智慧水库洪水调度系统开发，技术负责人

[7] 国家自然基金重大国际合作研究项目，河库连通下的特大水库群联合调度基础问题研究，技术骨干

授权专利

[1] 魏国振，丁伟，梁国华，何斌，唐榕，王猛，马杏，周惠成，一种考虑预报误差降低水库洪水起调水位的预报调度方法,国内发明专利

[2] 丁伟，李昱，张弛，周惠成，一种基于成本-效益分析的跨流域引水工程规模确定方法，国内发明专利（实现百万成果转化）

出版专著

[1] 梁团豪，王晓妮，李昱，丁伟，松花江流域骨干水库联合调度，大连理工大学出版社，2019年

[2] 李清清，马超，吴江，丁伟，何小聪，适应多维度用水需求的水库群供水调度技术，中国水利水电出版社，2021年