
基于新安江模型和BP神经网络的中小河流洪水模拟研究
李鑫, 刘艳丽, 朱士江, 王国庆, 金君良, 贺瑞敏, 刘翠善
基于新安江模型和BP神经网络的中小河流洪水模拟研究
Research on the Flood Simulation of Medium and Small Rivers Based on Xin'anjiang Model and BP Neural Network
为探讨更加符合中小河流域防洪要求的预报方法,并提高洪水预报精度,以屯溪流域为例,结合中小河流实际洪水预报要求,采用以洪峰合格率和峰现时间合格率为主要约束的非等权重的参数率定方法(即目标函数中径流深、洪峰流量、峰现时间合格率和确定性系数的权重分别为(1∶2∶2∶1)对新安江模型进行参数率定,并采用算术平均法耦合新安江模型和BP神经网络模型计算结果,以期提高洪水预报精度。结果表明:以洪峰合格率和峰现时间合格率为主要约束的率定方法的新安江模型是可行的,相对于传统等权重的率定方法,在洪峰和峰现时间预报方面更具优势,符合中小河流防洪要求;新安江模型对洪峰和峰现时间模拟较好,BP神经网络模型对洪峰和径流深的模拟表现较好,采用算数平均法耦合两者的模拟结果,可以提高洪水预报精度。
In order to explore the forecasting method which is more in line with the flood control requirements of the middle and small river basin and improve the accuracy of flood forecasting, taking Tunxi Basin as an example, combined with the actual flood forecasting requirements of the middle and small rivers, a non-equal-weight parameter calibration method with the qualified rate of flood peak and the qualified rate of flood peak onset time as the main constraints is adopted (i.e., the weights of runoff depth, flood peak discharge, qualified rate of peak onset time and deterministic coefficient in the objective function are respectively (1∶2∶2∶1) to calibrate the parameters of the Xin'anjiang River model, and use the arithmetic mean method to couple the calculation results of the Xin 'anjiang River model and BP neural network model so as to improve the accuracy of flood prediction. The results show that the method based on the main constraints of flood peak discharge and peak occurrence time is feasible in the flood prediction of Tunxi Basin. Compared with the traditional equal-weight method, the method has more advantages in the flood peak and peak occurrence time prediction, and meets the flood control requirements of small and medium-sized rivers. The Xin 'anjiang River model can simulate the flood peak and peak time well, and the BP neural network model can simulate the flood peak and runoff depth well. The arithmetic mean method is used to couple the simulation results of the two models can improve the accuracy of flood prediction.
中小河流 / 新安江模型 / BP神经网络模型 / 参数率定 / 耦合模型 {{custom_keyword}} /
small and medium rivers / Xin'anjiang model / BP neural network model / parameter calibration / coupling model {{custom_keyword}} /
表1 不同目标函数下新安江模型参数率定结果Tab.1 Parameter calibration results of Xin'anjiang model with different objective functions |
参数 | 物理意义 | 范围 | 目标函数 | |
---|---|---|---|---|
非等权重 | 等权重 | |||
K | 蒸散发折算系数 | 0.8~1.2 | 1.1 | 1.1 |
SM | 自由水蓄水容量 | 10~50 | 39 | 50 |
KG | 地下水出流系数 | 0~0.7 | 0.12 | 0.47 |
CG | 地下水消退系数 | 0.9~0.998 | 0.99 | 0.94 |
CI | 壤中流消退系数 | 0.01~0.9 | 0.83 | 0.25 |
CS | 河网水流消退系数 | 0.1~0.9 | 0.86 | 0.90 |
WM | 流域平均张力水容量 | 80~200 | 127 | 120 |
C | 深层蒸散发折减系数 | 0.1~0.2 | 0.18 | 0.18 |
B | 蓄水容量曲线方次 | 0.1~0.4 | 0.3 | 0.3 |
EX | 自由水蓄水容量曲线方次 | 1.0~1.5 | 1.5 | 1.5 |
l | 滞时 | - | 5 | 5 |
表2 不同目标函数率定结果下模拟结果统计表 (%)Tab.2 Statistical table of simulation results under calibration results of different objective functions |
率定方法 | RE 合格率 | RQ 合格率 | ΔH合格率 | 平均确定性系数 |
---|---|---|---|---|
非等权重法 | 85.37 | 100.00 | 95.12 | 0.82 |
等权重法 | 90.24 | 92.68 | 90.24 | 0.84 |
表3 模拟结果评价指标统计表Tab.3 Statistical table of evaluation indexes of simulation results |
方法 | 合格率/% | 平均绝对误差 | 平均确定性系数 | |||||
---|---|---|---|---|---|---|---|---|
RE | RQ | ΔH | RE | RQ | ΔH | |||
率定期 | 新安江模型 | 85.37 | 100.00 | 95.12 | 10.11 | 7.91 | 1.98 | 0.82 |
BP神经网络模型 | 100.00 | 97.56 | 92.68 | 3.05 | 8.58 | 3.63 | 0.90 | |
检验期 | 新安江模型 | 100.00 | 100.00 | 100.00 | 9.04 | 6.55 | 2.06 | 0.89 |
BP神经网络模型 | 100.00 | 100.00 | 100.00 | 2.27 | 6.14 | 3.00 | 0.93 | |
耦合模型计算结果 | 100.00 | 100.00 | 100.00 | 1.29 | 5.79 | 1.83 | 0.96 |
图2 两种模型模拟及耦合后流量过程线Fig.2 Flow process lines simulated and coupled by the two models |
表4 两种模型及耦合后评价指标结果统计表Tab.4 Statistical table of the results of the two models and the evaluation indexes after coupling |
洪水序号 | 模型 | RE /% | RQ /% | ΔH/h |
---|---|---|---|---|
20160420 | 新安江模型 | -5.09 | 6.28 | -2.00/±8.7 |
BP神经网络模型 | 0.93 | 4.93 | 6.00/±8.7 | |
耦合两种模型 | 0.27 | 6.48 | -2.00/±8.7 | |
20190515 | 新安江模型 | -17.63 | -0.76 | -1.00/±8.1 |
BP神经网络模型 | 1.45 | 7.08 | 5.00/±8.1 | |
耦合两种模型 | -0.58 | 6.35 | 1.00/±8.1 |
1 |
张建云.中国水文预报技术发展的回顾与思考[J].水科学进展,2010,21(4):435-443.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
2 |
刘志雨,杨大文,胡健伟.基于动态临界雨量的中小河流山洪预警方法及其应用[J].北京师范大学学报(自然科学版),2010,46(3):317-321.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
3 |
李红霞,王瑞敏,黄琦,等.中小河流洪水预报研究进展[J].水文,2020,40(3):16-23,50.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
4 |
刘志雨,刘玉环,孔祥意.中小河流洪水预报预警问题与对策及关键技术应用[J].河海大学学报(自然科学版),2021,49(1):1-6.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
5 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
6 |
刘志雨,侯爱中,王秀庆.基于分布式水文模型的中小河流洪水预报技术[J].水文,2015,35(1):1-6.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
7 |
霍文博,朱跃龙,李致家,等.新安江模型和支持向量机模型实时洪水预报应用比较[J].河海大学学报(自然科学版),2018,46(4):283-289.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
8 |
吴美玲,杨侃,张铖铖.基于KG-BP神经网络在秦淮河洪水水位预测中的应用[J].水电能源科学,2019(2):74-77,81.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
9 |
范火生.基于栅格的分布式新安江模型在三岔河流域的洪水预报研究[J].中国农村水利水电,2019(2):113-118.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
10 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
11 |
芮孝芳,凌哲,刘宁宁,等.新安江模型的起源及对其进一步发展的建议[J].水利水电科技进展,2012,32(4):1-5.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
12 |
刘佩瑶,郝振纯,王国庆,等.新安江模型和改进BP神经网络模型在闽江水文‘预报中的应用[J].水资源与水工程学报,2017,28(1):40-44
{{custom_citation.content}}
{{custom_citation.annotation}}
|
13 |
姚成,李致家,张珂,等.基于栅格型新安江模型的中小河流精细化洪水预报[J].河海大学学报(自然科学版),2021,49(1):19-25.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
14 |
宋晓猛,孔凡哲.新安江模型和人工神经网络的耦合应用[J].水土保持通报,2010,30(6):135-138,144.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
15 |
姚成,章玉霞,李致家,等.无资料地区水文模拟及相似性分析[J].河海大学学报(自然科学版),2013,41(2):108-113.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
16 |
包红军,李致家,王莉莉.降雨径流模拟神经网络模型及应用[J].西安建筑科技大学学报(自然科学版),2009,41(5):719-722.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
17 |
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.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
18 |
金保明,李芸,张烜,等.基于反向传播算法的崇阳溪流域洪水流量预报[J].水力发电,2017,43(8):22-25,38.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
19 |
杨洪.改进BP神经网络集成模型在径流预测中的应用[J].水资源与水工程学报,2014,25(3):213-219.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
20 |
王竹.半分布式耦合BP神经网络洪水预报模型研究[J].中国农村水利水电,2017(8):96-102.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
21 |
闫悦新,包为民.组合预报方法在洪水预报模型中的应用[J].水电能源科学,2013,31(10):47-49,73.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
22 |
肖洁,罗军刚,解建仓,等.变权组合预测模型在洪水预报中的应用[J].西北农林科技大学学报(自然科学版),2013,41(2):215-221.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |