Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)
土木与交通工程     
集合神经网络的洪水预报
江衍铭, 张建全, 明焱
1.浙江大学 建筑工程学院水文与水资源工程研究所 浙江 杭州 310058;
2.浙江大学 建筑工程学院建筑学系 浙江 杭州 310058
Flood forecasting by ensemble neural networks
CHIANG Yen ming, ZHANG Jian quan, MING Yan
1. Institute of Hydrology and Water Resources Engineering, Zhejiang University, Hangzhou 310058, China|
2. Department of Architecture, Zhejiang University, Hangzhou 310058, China
 全文: PDF(1247 KB)   HTML
摘要：

针对防洪减灾的问题,通过将集合预报概念应用于人工神经网络,综合考虑样本和参数等因素的影响,构建集合神经网络模型,以降低单一神经网络模型的不确定性.针对初始值扰动和样本重采样两方面分别产生集合成员,由简单平均和贝叶斯模型加权平均整合预报输出,构建龙泉溪流域预见期为1~3 h的集合洪水预报.结果表明,相对于单一神经网络,集合神经网络模型有效地提高预测的精度.从均方根误差上看,集合神经网络模型性能比单一神经网络模型提升了15%~35%.在众多集合策略中,以初始值扰动和简单平均操作最简单,模型预报输出有16%~32%的提升,重采样和贝叶斯模型加权平均的组合效果使预报精度改进了22%~35%.
Abstract:

Regarding the flood prevention, the concept of ensemble prediction was integrated into artificial neural networks by considering the data quality and initial weights to reduce the uncertainties of single neural network. The ensemble neural networks were built for 1~3 h ahead flood forecasting in Longquan River Basin, by generating the ensemble members through resampling and initialization methods and combining the model outputs through the arithmetic average and Bayesian model average (BMA). Results indicate that the ensemble neural networks significantly improve the flood forecasting accuracy as compared with the single neural network. The accuracy of ensemble neural network is about 15% to 35% higher than that of the single neural network in terms of root mean square error. Among various ensemble strategies, the combination of initialization and arithmetic average has simpler structure and less computational time; the improvement percentage rangs from 16% to 32%. The forecasts accuracy obtained from the combination of boosting and BMA are improved 22% to 35%.
出版日期: 2016-08-01
:     
基金资助:

教育部博士点新教师资助项目（J20131413）；浙江省教育厅一般资助项目（N20130036）.
通讯作者: 明焱,男,副教授.ORCID: 0000-0002-6889-9580     E-mail: mmmgmmm@vip.sina.com
作者简介: 江衍铭（1978—）,男,副教授,从事水文水资源等研究.ORCID: 0000-0002-6303-9303. E-mail: chiangym@zju.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

江衍铭, 张建全, 明焱. 集合神经网络的洪水预报[J]. 浙江大学学报(工学版), 10.3785/j.issn.1008-973X.2016.08.007.

CHIANG Yen ming, ZHANG Jian quan, MING Yan. Flood forecasting by ensemble neural networks. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 10.3785/j.issn.1008-973X.2016.08.007.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2016.08.007    或    http://www.zjujournals.com/eng/CN/Y2016/V50/I8/1471

［1］ 俞亭超,张土乔,毛根海,等.预测城市用水量的人工神经网络模型研究［J］.浙江大学学报:工学版,2004,38(9): 1156-1161．
YU Tingchao,ZHANG Tuqiao,MAO Genhai, et al. Study of artificial neural network model for forecasting urban water demand ［J］. Journal of Zhejiang University: Engineering Science, 2004, 38(9): 1156-1161．
［2］ 隋彩虹,徐宗学.人工神经网络模型在渭河下游洪水预报中的应用［J］.水文,2006,26(2): 38-42．
SUI Caihong,XU Zongxue. Application of ANN models in flood forecasting for the lower reach of the Weihe river ［J］.Journal of China Hydrology,2006,26 (2): 38-42．
［ 3 ］ CHIANG Y M, CHANG L C, CHANG F J. Comparison of staticfeedforward and dynamicfeedback neural networks for rainfallrunoff modeling ［J］. Journal of Hydrology, 2004, 290(3): 297-311．
［4］ 徐伟,董增川,付晓花,等.基于BP人工神经网络的河流生态健康预警［J］.河海大学学报:自然科学版,2015,43(1): 54-59．
XU Wei,DONG Zengchuan,FU Xiaohua, et al． Early warning of river ecosystem health based on BP artificial neural networks ［J］.Journal of  Hohai University ：Natural Sciences, 2015, 43(1): 54-59．
［5］ 胡健伟,周玉良,金菊良.BP神经网络洪水预报模型在洪水预报系统中的应用［J］.水文,2015,35(1): 20-25．
HU Jianwei, ZHOU Yuliang, JIN Juliang. Flood forecasting model on BP networks and its application in flood forecasting systems ［J］.Journal of China  Hydrology, 2004, 290(3): 297-311．
［6］ 吕庆平,罗坚,朱坤,等.基于人工神经网络的热带气旋路径预报试验［J］.广东气象,2008,31(1): 15-18．
LU Qingping, LUO Jian, ZHU Kun, et al. Experiments on predicting tracks of tropical cyclones based on artificial neural network ［J］. Guangdong Meteorology, 2008, 31(1): 15-18．
［7］ 孙军波,钱燕珍,陈佩燕,等.登陆台风站点大风预报的人工神经网络方法［J］.气象,2010, 36(9): 81-86．
SUN Junbo,QIAN Yanzhen,CHEN Peiyan, et al. The artificial neural network method on the station wind in landfall typhoon ［J］. Meteorological Monthly, 2010, 39(9): 81-86.
［8］ 包红军,赵琳娜.基于集合预报的淮河流域洪水预报研究［J］.水利学报,2012,43(2): 216-224.
BAO Hongjun,ZHAO Linna. Flood forecast of Huaihe river based on TIGGE ensemble predictions ［J］.Journal of Hydraulic Engineering, 2012, 43(2): 216-224．
［9］ 孙照渤,谭桂容,赵振国,等.中国东部夏季雨型的人工神经网络集合预测［J］.大气科学学报,2013,36(1): 16.
SUN Zhaobo,TAN Guirong,ZHAO Zhenguo, et al. Ensemble prediction of summer rainfall patterns over eastern China based on artificial neural networks ［J］.Journal of Nanjing Institute of Meteorology, 2013, 36(1): 16．
［10］ KASIVISWANATHAN K S, CIBIN R, SUDHEER K P, et al. Constructing prediction interval for artificial neural network rainfall runoffmodels based on ensemble simulations ［J］. Journal of Hydrology, 2013, 499: 275-288．
［ 11 ］ ElSHAFIE A, NAJAH A, ALSULAMI H M, et al. Optimized neural network prediction model for potential evapotranspiration utilizing ensemble procedure ［J］. Water Resources Management, 2014, 28(4): 947-967．
［12］ ZAIER I, SHU C, OUAR-DA T B M J, et al. Estimation of ice thickness on lakes using artificial neural network ensembles ［J］. Journal of Hydrology, 2010, 383(3): 330-340.
［13］ OUARDA T B M J, SHU C. Regional lowflow frequency analysis using single and ensemble artificial neural networks ［J］. Water Resources Research, 2009, 45(11): 114122．
［14］ ARAGHINEJAD S, AZMI M, KHOLGHI M. Application of artificial neural network ensembles in probabilistic hydrological forecasting ［J］. Journal of Hydrology, 2011, 407(14): 94-104．
［15］ MONMOUTH H, HARRIS DRUCKER. Improving regressors using boosting techniques ［C］∥ Proceedings of the Fourteenth International Conference on Machine Learning. San Francisco: Morgan Kaufmann Publishers Inc. 1997: 107-115．
［16］ 杜新忠,李叙勇,王慧亮,等.基于贝叶斯模型平均的径流模拟及不确定性分析［J］.水文,2014,34(3): 6-10.
DU Xinzhong,LI Xuyong,WANG Huiliang, et al. Multimodel Ensemble runoff simulation based on bayesian model averaging method and model structure uncertainty analysis ［J］.Journal of China Hydrology, 2014, 34(3): 6-10．
［17］ 董磊华,熊立华,万民.基于贝叶斯模型加权平均方法的水文模型不确定性分析［J］.水利学报,2011,42(9): 1065-1074．
DONG Leihua,XIONG Lihua,WAN Min．Uncertainty analysis of hydrological modeling using the Bayes Model Averaging Method ［J］.Journal of Hydraulic Engineering,2011, 42(9): 1065-1074.
［18］ SEIBERT J. On the need for benchmarks in hydrological modeling ［J］. Hydrological Processes, 2001, 15(6): 1063-1064．
［19］ HERR H D, KRYSZTOFOWICZ R. Bayesian ensemble forecast of river stages and ensemble size requirements ［J］. Journal of Hydrology, 2010, 387(13):151-164．
[1] 董凯, 赖俊英, 钱晓倩, 詹树林, 阮方. 夏热冬冷地区居住建筑水平式外遮阳节能效果[J]. 浙江大学学报(工学版), 2016, 50(8): 1431-1437.
[2] 李佳琦, 范利武, 俞自涛. 超亲水表面在淬火冷却过程中的沸腾传热特性[J]. 浙江大学学报(工学版), 2016, 50(8): 1493-1498.
[3] 钟崴, 彭梁, 周永刚, 徐剑, 从飞云. 基于小波包分析和支持向量机的锅炉结渣诊断[J]. 浙江大学学报(工学版), 2016, 50(8): 1499-1506.
[4] 夏玉峰, 任莉, 叶彩红, 王力. 基于RSM的立柱加强板定位布局多目标优化[J]. 浙江大学学报(工学版), 2016, 50(8): 1600-1607.
[5] 李林玉, 吴张华, 余国瑶, 戴巍, 罗二仓. 直线压缩机电声转换特性的实验[J]. 浙江大学学报(工学版), 2016, 50(8): 1529-1536.
[6] 曲巍崴, 唐伟, 毕运波, 李少波, 罗水均. 避免强迫装配和提升效率的预连接工艺规划[J]. 浙江大学学报(工学版), 2016, 50(8): 1561-1569.
[7] 胡小东, 顾临怡, 张范蒙. 应用于数字变量马达的高速开关阀[J]. 浙江大学学报(工学版), 2016, 50(8): 1551-1560.
[8] 杨姝, 刘国平, 亓昌, 王大志. 金属空心球梯度泡沫结构抗冲击特性仿真与优化[J]. 浙江大学学报(工学版), 2016, 50(8): 1593-1599.
[9] 杨章, 童根树, 张磊. 对称布置2根单侧加劲肋的有效刚度[J]. 浙江大学学报(工学版), 2016, 50(8): 1446-1455.
[10] 蒋翔, 童根树, 张磊. 耐火钢-混凝土组合梁抗火性能试验[J]. 浙江大学学报(工学版), 2016, 50(8): 1463-1470.
[11] 单华峰, 夏唐代, 俞峰, 胡军华, 潘金龙. 地下增层开挖托换桩的屈曲稳定临界荷载分析[J]. 浙江大学学报(工学版), 2016, 50(8): 1425-1430.
[12] 辜天来,张帅,郑耀. 咽式进气道/等直隔离段的反压特性[J]. 浙江大学学报(工学版), 2016, 50(7): 1418-1424.
[13] 程时伟, 陆煜华, 蔡红刚. 移动设备眼动跟踪技术[J]. 浙江大学学报(工学版), 2016, 50(6): 1160-1166.
[14] 郑成志, 高金良, 何文杰. 基于FastICA算法的物理漏损流量分析模型[J]. 浙江大学学报(工学版), 2016, 50(6): 1031-1039.
[15] 黄家辉, 冯冬芹. 广义收益信息物理系统脆弱性评估方法[J]. 浙江大学学报(工学版), 2016, 50(6): 1119-1125.