基于弹性梯度下降算法的BP神经网络降雨径流预报模型

doi:10.6040/j.issn.1672-3961.0.2019.504

摘要/Abstract

摘要：

运用反向传播(back propagation, BP)的改进算法弹性梯度下降算法,选择崇阳溪上游流域1997—2014年的14场降雨径流过程,以流域内洋庄、吴边、大安、坑口、岭阳、岚谷6个雨量站的实测降雨量和武夷山水文站的前期流量资料为输入,武夷山水文站相应流量为输出,建立弹性梯度下降算法的BP神经网络降雨径流预报模型,采用7场降雨径流过程对模型进行检验。结果表明,与传统的反向传播算法相比,该模型所需的参数较少,运算速度显著提高,模型的预报精度满足要求,可以为防汛部门预测洪水提供依据。

关键词: 弹性梯度下降法, BP神经网络, 降雨径流, 预报模型

Abstract:

The improved elastic gradient descent algorithm of back propagation was used, and 14 rainfall runoff processes from 1997 to 2014 in the upper reaches of Chongyang River were selected. The back propagation (BP) neural network rainfall-runoff forecasting model of the elastic gradient descent algorithm was established, which took the measured rainfall of six rainfall stations in Yangzhuang, Wubian, Da′an, Kengkou, Lingyang, and Langu in the basin and the preliminary flow data of Wuyishan Hydrological Station as inputs, and selected the corresponding flow of Wuyishan Hydrological Station as output. The 7-rainfall runoff process was used to test the model, the test results showed that the proposed method required fewer parameters and had higher operation speed than the traditional back propagation algorithm. The prediction accuracy of the model could meet the requirements, and provide the basis for flood control and disaster reduction.

Key words: elastic gradient descent algorithm, BP neural network, rainfall runoff, forecasting model

中图分类号:

TV124

金保明,卢光毅,王伟,杜伦阅. 基于弹性梯度下降算法的BP神经网络降雨径流预报模型[J]. 山东大学学报 (工学版), 2020, 50(3): 117-124.

Baoming JIN,Guangyi LU,Wei WANG,Lunyue DU. Research on BP neural network rainfall runoff forecasting model based on elastic gradient descent algorithm[J]. Journal of Shandong University(Engineering Science), 2020, 50(3): 117-124.

图/表 7

图1

图2

图3

图4

表1

表2

表3

参考文献 18

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.
	ZHANG Jianyun . Review and reflection on China's hydrological forecasting techniques[J]. Advances in Water Science, 2010, 21 (4): 435- 443.
2	包红军, 王莉莉, 李致家, 等. 基于Holtan产流的分布式水文模型[J]. 河海大学学报(自然科学版), 2016, 44 (4): 340- 346.
	BAO Hongjun , WANG Lili , LI Zhijia , et al. A distributed hydrological model based on Holtan runoff generation theory[J]. Journal of Hohai University(Natural Sciences), 2016, 44 (4): 340- 346.
3	何昳颖, 陈晓宏, 张云, 等. BP人工神经网络在小流域径流模拟中的应用[J]. 水文, 2015, (5): 35- 40. doi: 10.3969/j.issn.1000-0852.2015.05.007
	HE Yiying , CHEN Xiaohong , ZHANG Yun , et al. Application of BP neural network model in runoff simulating for small watershed[J]. Journal of China Hydrology, 2015, (5): 35- 40. doi: 10.3969/j.issn.1000-0852.2015.05.007
4	王胜刚, 张莹, 徐应涛, 等. 基于打洞函数法的BP神经网络水文预报方法[J]. 运筹学学报, 2011, 15 (4): 45- 54. doi: 10.3969/j.issn.1007-6093.2011.04.005
	WANG Shenggang , ZHANG Ying , XU Yingtao , et al. The BP neural network hydrological forecasting algorithm based on tunneling function method[J]. Operations Research Transactions, 2011, 15 (4): 45- 54. doi: 10.3969/j.issn.1007-6093.2011.04.005
5	NOURANI V . An emotional ANN (EANN) approach to modeling rainfall-runoff process[J]. Journal of Hydrology, 2017, 544, 267- 277. doi: 10.1016/j.jhydrol.2016.11.033
6	杜拉, 纪昌明, 李荣波, 等. 基于小波-BP神经网络的贝叶斯概率组合预测模型及其应用[J]. 中国农村水利水电, 2015, (7): 50- 53. doi: 10.3969/j.issn.1007-2284.2015.07.012
	DU La , JI Changming , LI Rongbo , et al. Wavelet-BP neural network Bayesian probabilistic combination forecasting model and its application[J]. China Rural Water and Hydropower, 2015, (7): 50- 53. doi: 10.3969/j.issn.1007-2284.2015.07.012
7	SEDKIA A , MAZOUDI E E . Evolving neural network using real coded genetic algorithm for daily rainfall-runoff forecasting[J]. Expert Systems with Applications, 2009, (36): 4523- 4527.
8	WANG Jianjin , PENG Shi . Application of BP neural network algorithm in tradition hydrological model for flood forecasting[J]. Water, 2017, 9 (1): 1- 16.
9	INAN C A , CANOGLU M C , KURTULUS B . A new artificial neural network model for the prediction of the rainfall-runoff relationship for La Chartreux Spring, France[J]. Fresenius Environmental Bulletin, 2018, 27 (11): 7354- 7363.
10	马江琦.基于奇异值分解和特征融合的人脸识别[D].济南:山东大学, 2016.
	MA Jiangqi. Face recognition algorithm based on singular value decomposition and feature fusion[D]. Jinan: Shandong University, 2016.
11	ZHANG R , LEE E . Human face recognition used improved back-propagation (BP) neural network[J]. Journal of Korea Multimedia Society, 2018, 21 (4): 471- 477.
12	NARESH BABU , EDLA DAMODAR-REDDY . New algebraic activation function for multi-layered feed forward neural networks[J]. IETE Journal of Research, 2017, 63 (1): 71- 79. doi: 10.1080/03772063.2016.1240633
13	邢茂琳, 周新志. 都江堰来水预测中BP网络训练方法的研究[J]. 微计算机信息, 2009, (34): 149- 150.
	XING Maolin , ZHOU Xinzhi . Study of BP training method based on Dujiangyan water prediction[J]. Microcomputer Information, 2009, (34): 149- 150.
14	熊海晶.基于神经网络和小波分析的降水预报研究[D].南京:南京大学, 2012.
	XIONG Haijing. Study on the precipitation forecast based on neural network and wavelet analysis[D]. Nanjing: Nanjing University, 2012.
15	RIEDMILLER M , BRAUN H . A direct adaptive method for faster back propagation learning: the RPROP algorithm[J]. International Conference on Neural Networks, San Francisco, USA: IEEI, 1993, 1, 586- 591.
16	袁曾任. 人工神经元网络及其应用[M]. 北京: 清华大学出版社, 1999.
17	赵振宇, 徐用懋. 模糊理论和神经网络的基础与应用[M]. 南宁: 广西科学技术出版社, 1996.
18	中华人民共和国国家质量监督检验检疫总局,中国国家标准化管理委员会.水文情报预报规范: GB/T 22482—2008[S].北京:中国标准出版社, 2008.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

洪水场次	绝对误差绝对值平均值/(m³·s^-1)	相对误差绝对值平均值/%	确定性系数
1998·06·14	22.6	3.4	0.988
1999·06·17	14.5	9.2	0.997
2003·06·25	16.2	8.1	0.984
2005·06·19	35.2	7.6	0.973
2007·06·14	27.8	15.7	0.978
2008·07·19	43.6	10.7	0.991
2013·05·27	12.0	7.9	0.986

洪水场次	峰型	实测流量/(m³·s^-1)	预报流量/(m³·s^-1)	绝对误差/(m³·s^-1)	相对误差/%
1998·06·14	主峰	3 080	2 929	-151	-4.9
1998·06·14	次峰	2 710	2 689	-21	-0.8
1999·06·17	主峰	1 170	1 192	22	1.9
1999·06·17	次峰	988	975	-13	1.3
2003·06·25	主峰	797	872	75	9.4
2005·06·19	主峰	1 250	1 277	27	2.2
2005·06·19	次峰	1 180	1 244	64	5.4
2007·06·14	主峰	887	943	56	6.3
2008·07·19	主峰	2 210	2 044	-166	7.5
2013·05·27	主峰	474	425	-49	9.3
平均值	-	1 475	1 459	64	5.0

洪水场次	实测径流深/mm	预报径流深/mm	绝对误差/mm	相对误差/%
1998·06·14	299.6	301.0	1.4	0.5
1999·06·17	98.7	101.0	1.6	1.6
2003·06·25	83.9	88.1	4.2	5.0
2005·06·19	116.2	119.1	2.9	2.5
2007·06·14	36.1	39.2	3.1	8.6
2008·07·19	85	87.3	2.3	2.7
2013·05·27	23.2	23.9	0.9	3.1
平均值	106.1	108.5	2.3	3.4

[1]	郑店坤,许同乐,尹召杰,孟庆民. 改进PSO-BP神经网络对尾矿坝地下水位的预测方法[J]. 山东大学学报 (工学版), 2019, 49(3): 108-113.
[2]	杨亚楠,夏斌,谢楠,袁文浩. 基于BP神经网络和多元Taylor级数的混合定位算法[J]. 山东大学学报 (工学版), 2019, 49(1): 36-40.
[3]	刘杰, 杨鹏, 吕文生, 刘阿古达木, 刘俊秀. 基于气象因素的PM_2.5质量浓度预测模型[J]. 山东大学学报(工学版), 2015, 45(6): 76-83.
[4]	姚福安,庞向坤,焦营营,王忠林,张锡满 . 基于三色法和BP神经网络的回转窑温度检测[J]. 山东大学学报(工学版), 2008, 38(2): 61-65 .
[5]	高小伟,蒋晓芸 . BP神经网络在入侵检测系统中的应用及优化[J]. 山东大学学报(工学版), 2006, 36(6): 107-110 .