Heuristic Methods for Reservoir Monthly Inflow Forecasting: A Case Study of Xinfengjiang Reservoir in Pearl River, China

被引:37
作者
Cheng, Chun-Tian [1 ]
Feng, Zhong-Kai [1 ]
Niu, Wen-Jing [1 ]
Liao, Sheng-Li [1 ]
机构
[1] Dalian Univ Technol, Inst Hydropower & Hydroinformat, Dalian 116024, Peoples R China
关键词
ARTIFICIAL NEURAL-NETWORKS; SUPPORT VECTOR REGRESSION; GENETIC ALGORITHM; HYBRID MODEL; TERM FORECAST; PREDICTION; PERFORMANCE; CALIBRATION; DISCHARGE; MACHINE;
D O I
10.3390/w7084477
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Reservoir monthly inflow is rather important for the security of long-term reservoir operation and water resource management. The main goal of the present research is to develop forecasting models for the reservoir monthly inflow. In this paper, artificial neural networks (ANN) and support vector machine (SVM) are two basic heuristic forecasting methods, and genetic algorithm (GA) is employed to choose the parameters of the SVM. When forecasting the monthly inflow data series, both approaches are inclined to acquire relatively poor performances. Thus, based on the thought of refined prediction by model combination, a hybrid forecasting method involving a two-stage process is proposed to improve the forecast accuracy. In the hybrid method, the ANN and SVM are, first, respectively implemented to forecast the reservoir monthly inflow data. Then, the processed predictive values of both ANN and SVM are selected as the input variables of a newly-built ANN model for refined forecasting. Three models, ANN, SVM, and the hybrid method, are developed for the monthly inflow forecasting in Xinfengjiang reservoir with 71-year discharges from 1944 to 2014. The comparison of results reveal that three models have satisfactory performances in the Xinfengjiang reservoir monthly inflow prediction, and the hybrid method performs better than ANN and SVM in terms of five statistical indicators. Thus, the hybrid method is an efficient tool for the long-term operation and dispatching of Xinfengjiang reservoir.
引用
收藏
页码:4477 / 4495
页数:19
相关论文
共 45 条
[1]   Water level forecasting through fuzzy logic and artificial neural network approaches [J].
Alvisi, S ;
Mascellani, G ;
Franchini, M ;
Bárdossy, A .
HYDROLOGY AND EARTH SYSTEM SCIENCES, 2006, 10 (01) :1-17
[2]  
Bazartseren B, 2003, NEUROCOMPUTING, V55, P439, DOI [10.1016/S0925-2312(03)00388-6, 10.1016/S0925-2312(03)00388-5]
[3]   A hybrid model coupled with singular spectrum analysis for daily rainfall prediction [J].
Chau, K. W. ;
Wu, C. L. .
JOURNAL OF HYDROINFORMATICS, 2010, 12 (04) :458-473
[4]   Comparison of several flood forecasting models in Yangtze river [J].
Chau, KW ;
Wu, CL ;
Li, YS .
JOURNAL OF HYDROLOGIC ENGINEERING, 2005, 10 (06) :485-491
[5]   Intelligent manipulation and calibration of parameters for hydrological models [J].
Chen, W. ;
Chau, K. W. .
INTERNATIONAL JOURNAL OF ENVIRONMENT AND POLLUTION, 2006, 28 (3-4) :432-447
[6]  
Cheng CT, 2005, LECT NOTES COMPUT SC, V3612, P1152
[7]   Combining a fuzzy optimal model with a genetic algorithm to solve multi-objective rainfall-runoff model calibration [J].
Cheng, CT ;
Ou, CP ;
Chau, KW .
JOURNAL OF HYDROLOGY, 2002, 268 (1-4) :72-86
[8]   Using a hybrid genetic algorithm-simulated annealing algorithm for fuzzy programming of reservoir operation [J].
Chiu, Yu-Chen ;
Chang, Li-Chiu ;
Chang, Fi-John .
HYDROLOGICAL PROCESSES, 2007, 21 (23) :3162-3172
[9]   Daily reservoir inflow forecasting using artificial neural networks with stopped training approach [J].
Coulibaly, P ;
Anctil, F ;
Bobée, B .
JOURNAL OF HYDROLOGY, 2000, 230 (3-4) :244-257
[10]   Improving daily reservoir inflow forecasts with model combination [J].
Coulibaly, P ;
Haché, M ;
Fortin, V ;
Bobée, B .
JOURNAL OF HYDROLOGIC ENGINEERING, 2005, 10 (02) :91-99