河流径流预测作为水库调度和发电的重要前提,其预测精度直接影响水利工程的综合效益。基于径流历史数据,本文针对其波动和随机性提出一种小波分析-支持向量机(SVM)特征分类组合预测模型。该模型首先利用小波分解提取原始径流序列的高低频能量谱作为SVM样本标记,并对原始序列进行特征分类,分为“平稳型”和“突变型”序列,对应不同类型序列的小波近似信号和细节信号分别采用自回归和滑动平均模型(ARMA)和BP神经网络模型进行预测,再重构各序列预测结果。以2013年1月至2017年11月宜昌站三峡入库径流为例,最后采用平均绝对百分比误差(MAPE)、均方根误差(RMSE)、希尔不等式系数(TIC)作为模型评价指标。结果表明：在3个评价指标下相比ARMA和BP神经网络,文中模型在枯水期MAPE为1.16%,预测精度提高6%,在丰水期MAPE为1.84%,预测精度提高8%,由此验证了本文所提模型的精确性,且预测序列平滑,模型具有更好预测稳定性。
Abstract
As an important prerequisite for reservoir operation and power generation, the prediction accuracy of river runoff has a direct impact on the comprehensive benefits of water conservancy projects. Based on the historical data of runoff, this paper proposes a wavelet analysis support vector machine (SVM) feature classification combined forecasting model for its volatility and randomness. Firstly, the wavelet decomposition is used to extract the high and low frequency energy spectrum of the original runoff sequence as the SVM sample mark, and the original sequence is classified by feature, dividing into stationary and abrupt sequences, the wavelet approximation signals and the detail signals, corresponding to different types of sequences, are predicted by auto-regressive moving average model (ARMA) and BP neural network model respectively, then the prediction results of each sequence are reconstructed. Finally, the Mean Absolute Percentage Error (MAPE), the Root Mean Square Error (RMSE) and the Theil Inequality Coefficient (TIC) are used as the evaluation indexes of the model. The results show that: under the 3 evaluation indexes, the proposed model is better than the ARMA and BP neural network models, and it has better prediction stability.
关键词
径流预测 /
小波分解 /
支持向量机 /
自回归和滑动平均模型 /
神经网络 /
特征分类
{{custom_keyword}} /
Key words
runoff prediction /
wavelet decomposition /
support vector machine /
auto-regressive moving average model /
artificial neural networks /
feature classification
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Antonetti M, Scherrer S, Kienzler P M, et al. Process‐based hydrological modelling: The potential of a bottom‐up approach for runoff predictions in ungauged catchments [J]. Hydrological Processes, 2017.
[2] 孟二浩, 黄生志, 黄强,等. 融合大气环流异常因子的径流预报研究[J]. 水力发电学报, 2017, 36(8):34-42.
Meng Erhao, Huang Shengzhi, Huang Qiang, et al. Runoff prediction incorporating anomalous atmospheric circulation factors [J]. Journal of Hydroelectric Engineering , 2017, 36 (8): 34-42.
[3] 于瑞宏, 张宇瑾, 张笑欣,等. 无测站流域径流预测区域化方法研究进展[J]. 水利学报, 2016, 47(12):1528-1539.
Yu Ruihong, Zhang Yujin, Zhang Xiaoxin, et al. Review of regionalization methods on streamflow prediction in ungauged basins [J]. Journal of Hydraulic Engineering, 2016, 47 (12): 1528-1539.
[4] 尹鑫卫, 李晓玲, 康燕霞,等. 基于SCS-CN模型的沟垄微型集雨系统径流预测[J]. 生态学杂志, 2015, 34(12):3502-3508.
Yin Xinwei, Li Xiaoling, Kang Yanxia, et al. Prediction of runoff in ridge-furrow rainwater harvesting system based on SCS-CN model [J]. Chinese Journal of Ecology, 2015, 34 (12): 3502-3508.
[5] 刘国东, 丁晶. BP网络用于水文预测的几个问题探讨[J]. 水利学报, 1999(1):65-70.
Liu Guodong, Ding Jing. Discussion on problems of BP neural networks applied to hydrological prediction [J]. Journal of Hydraulic Engineering, 1999 (1): 65-70.
[6] Shoaib M, Shamseldin A Y, Melville B W, et al. A comparison between wavelet based static and dynamic neural network approaches for runoff prediction[J]. Journal of Hydrology, 2016, 535:211-225.
[7] 周娅, 郭萍, 古今今. 基于BP神经网络的概率径流预测模型[J]. 水力发电学报, 2014, 33(2):45-50.
Zhou ya, Guo Ping, Gu Jinjin. Probabilistic runoff forecasting model based on BP artificial neural network [J]. Journal of Hydroelectric Engineering, 2014, 33 (2): 45-50.
[8] Vapnik V N. The nature of statistical learning theory[M]. Springer, 2000.
[9] 叶碎高, 彭勇, 周惠成. 基于PSO参数辨识SVM的中长期径流预测研究[J]. 大连理工大学学报, 2011, 51(1):115-120.
Ye Suigao, Peng Yong, Zhou Huicheng. Research on support vector machine parameter identification method for mid and long-term runoff forecast based on particle swarm optimization algorithm [J]. Journal of Dalian University of Technology, 2011, 51 (1): 115-120.
[10] Meng X, Yin M, Ning L, et al. A threshold artificial neural network model for improving runoff prediction in a karst watershed[J]. Environmental Earth Sciences, 2015, 74(6):1-10.
[11] 卫太祥, 马光文, 黄炜斌. 基于惩罚加权支持向量机回归的径流预测模型[J]. 水力发电学报, 2012, 31(6):35-38.
Wei Taixiang, Ma Guangwen, Huang Weibin. Runoff forecast based on weighted support vector machine regression model [J]. Journal of Hydroelectric Engineering, 2012, 31 (6): 35-38.
[12] 聂敏, 刘志辉, 刘洋,等. 基于PCA和BP神经网络的径流预测[J]. 中国沙漠, 2016, 36(4):1144-1152.
Nie Min, Liu Zhihui, Liu Yang, et al. Runoff forecast based on principal component analysis and BP neural network [J]. Journal of Desert Research , 2016, 36 (4): 1144-1152.
[13] 李娇, 姜明媛, 孙文超,等. 基于BP神经网络的泉州市山美水库降雨径流模拟研究[J]. 北京师范大学学报(自然科学版), 2013, 49(2):170-174.
Li Jiao, Jiang Mingyuan, Sun Wenchao, et al. Rainfall-runoff simulation of ShanMei reservoir in QuanZhou city on BP neural networks [J]. Journal of Beijing Normal University(Natural Science), 2013, 49 (2): 170-174.
[14] 黄巧玲, 粟晓玲, 杨家田. 基于小波分解的日径流支持向量机回归预测模型[J]. 西北农林科技大学学报(自然科学版), 2016, 44(4):211-217.
Huang Qiaoling, Su Xiaoling, Yang Jia Tian. Wavelet based support vector machine regression model for daily runoff prediction [J]. Journal of Northwest A&F University (Natural Science Edition), 2016, 44 (4): 211-217.
[15] Patil S K, Valunjkar S S. Utility of Coactive Neuro-Fuzzy Inference System for Runoff Prediction in Comparison with Multilayer Perception[J]. International Journal of Engineering Research, 2016, 5(1):156-160.
[16] 纪昌明, 李荣波, 张验科,等. 基于小波分解的投影寻踪自回归组合模型及其在年径流预测中的应用[J]. 水力发电学报, 2015, 34(7):27-35.
Ji Changming, Li Rongbo, Zhang Ke, et al. Projection pursuit autoregression model based on wavelet decomposition and its application in annual runoff prediction [J]. Journal of Hydroelectric Engineering, 2015, 34 (7): 27-35.
[17] 王秀杰, 封桂敏, 耿庆柱. 小波分析组合模型在日径流预测中的应用研究[J]. 自然资源学报, 2014(5):885-893.
Wang Xiujie, Feng Guimin, Geng Qinzhu. Application Research on Combined Models Based on WaveletAnalysis in Prediction of Daily Runoff [J]. Journal of Natural Resources, 2014 (5): 885-893.
[18] Shi J, Liu Y, Yang Y, et al. Short-term wind power prediction based on wavelet transform-support vector machine and statistic characteristics analysis[C]// Industrial and Commercial Power Systems Technical Conference. IEEE, 2011:1136-1141.
[19] Hartman E, Keeler J D, Kowalski J M. Layered Neural Networks with Gaussian Hidden Units as Universal Approximations [J]. Neural Computation, 1990, 2(2):210-215.
[20] 王秀杰, 练继建, 费守明. 基于小波消噪的混沌神经网络径流预报模型[J]. 水力发电学报, 2008, 27(5):37-40.
Wang Xiujie, Lian Jijian, Fei Shou Ming. The chaotic neural network model of runoff forecast based on wavelet de-noising [J]. Journal of Hydroelectric Engineering, 2008, 27 (5): 37-40.
[21] 易丹辉. 数据分析与EViews应用.第2版[M]. 中国人民大学出版社, 2014.
[22] 谢中华. MATLAB统计分析与应用[M]. 北京航空航天大学出版社, 2010
PDF(2003 KB)
鄂公网安备 42010602001274号 鄂ICP备14015464号-1