基于机器集成学习的中长期径流预报研究
Research on Medium and Long-Term Runoff Forecasting Based on Machine Integrated Learning

DOI: 10.12677/JWRR.2021.101005, PP. 44-52

Keywords: 径流预报，集成学习，机器学习，锦屏一级水库
Runoff Forecast, Ensemble Learning, Machine Learning, The Jinping-1 Reservoir

Full-Text   Cite this paper   Add to My Lib

Abstract:

中长期径流预报对水库优化调度及水资源优化开发利用都有着重要的意义。首先采用基于Boosting算法的梯度提升回归树(Gradient Boosting Decision Tree, GBRT)和极端梯度提升树(Extreme Gradient Boosting, XGBoost)、基于Bagging算法的随机森林(Random Forest, RF)和极端随机树(Extreme Random Tree, ET)四种算法作为预报模型对锦屏一级水库月平均入库流量序列进行预报，并对预测结果进行对比分析。结果显示，RF预测效果最差，XGBoost预测效果最好。进一步选用其中预测效果较好的三个方法ET、XGBoost、GBRT作为初级学习器，以Logistic回归作为次学习器，进行Stacking集成学习预测。结果表明，Stacking集成学习的预测效果要优于单一模型中预测效果最好的XGBoost，其预测值的结果和实测值更为接近，为中长期径流预报提供了新思路。
Medium and long-term runoff forecast is of great significance to the optimal operation of reservoirs, development and utilization of water resources. Firstly, the gradient boosting decision tree (GBRT) and extreme gradient boosting (XGBoost) based on boosting algorithm are selected. There is also random forest (RF) and extreme random tree (ET) based on bagging algorithm. These four algorithms are used as forecasting models to forecast the average monthly inflow of the Jinping-I Reservoir, and then the prediction results are analyzed and compared. The results showed that the RF prediction was the worst, and XGBoost was the best. Then, the three methods with better prediction effect are ET, XGBoost and GBRT as primary learners, logistic regression as secondary learners, and stacking ensemble learning to predict. The first mock exam results show that the prediction result of Stacking ensemble learning is better than that of XGBoost with the best prediction result in a single model. The predicted value is closer to the measured value, which provides a new idea for medium and long-term runoff forecast.

References

[1]	朱双. 流域中长期水文预报与水资源承载力评价方法研究[D]: [博士学位论文]. 武汉: 华中科技大学, 2017. ZHU Shuang. Studies on watershed long-term hydrological forecast and evaluation method of water resources carrying capacity. Ph.D. Thesis, Wuhan: Huazhong University of Science & Technology. 2017. (in Chinese)
[2]	KIRK M. p ython机器学习实践: 测试驱动的开发方法[M]. 北京: 机械工业出版社, 2018. KIRK M. Thoughtful machine learning with Python: A test-driven approach. Beijing: China Machine Press, 2018. (in Chi-nese)
[3]	李伶杰, 王银堂, 胡庆芳, 刘定忠, 张安富, 巴亚荃. 基于随机森林与支持向量机的水库长期径流预报[J]. 水利水运工程学报, 2020(4): 33-40. LI Lingjie, WANG Yintang, HU Qingfang, LIU Dingzhong, ZHANG Anfu, and BAYAQUAN. Long-term reservoir runoff forecast based on random forest and support vector machine. Journal of Water Resources and Water Transport Engineering, 2020(4): 33-40. (in Chinese)
[4]	左岗岗. 基于机器学习的渭河流域径流预测系统研究[D]: [硕士学位论文]. 西安: 西安理工大学, 2017. ZUO Ganggang. The research of WEI River runoff prediction system based on machine learning. Master’s Thesis, Xi’an: Xi’an University of Technology, 2017. (in Chinese)
[5]	许斌, 杨凤根, 郦于杰. 两类集成学习算法在中长期径流预报中的应用[J]. 水力发电, 2020, 46(4): 21-24+34. XU Bin, YANG Fenggen, and LI Yujie. Application of two types of integrated learning algorithms in mid- and long-term runoff forecasting. Hydropower, 2020, 46(4): 21-24+34. (in Chinese)
[6]	胡丹. 面向视觉跟踪的深度学习模型设计与优化研究[D]: [博士学位论文]. 西安: 西北工业大学, 2017. HU Dan. Model design and optimization of deep learning for visual tracking. Ph.D. Thesis, Xi’an: Northwestern Polytechnical University, 2017. (in Chinese)
[7]	陈宏, 邓芳明, 吴翔, 付智辉. 基于梯度提升决策树的电力电子电路故障诊断[J]. 测控技术, 2017, 36(5): 9-12+20. CHEN Hong, DENG Fangming, WU Xiang, and FU Zhihui. Power electronic circuit fault diagnosis based on gradient boosting decision tree. Measurement and Control Technology, 2017, 36(5): 9-12+20. (in Chinese)
[8]	侯舒凯. 基于集成学习方法的MINIST手写数字识别[J]. 通讯世界, 2018(8): 236-237. HOU Shukai. MINIST handwritten digit recognition based on integrated learning method. Communication World, 2018(8): 236-237. (in Chinese)
[9]	宋俊杰. 三峡流域中长期径流预报模型精度评定综合分析及优化方法研究[D]: [硕士学位论文]. 武汉: 华中科技大学, 2013. SONG Junjie. Comprehensive analysis and optimization method of the accuracy assessment of the medium and long-term runoff forecast model in the Three Gorges Basin. Master’s Thesis, Wuhan: Huazhong University of Science and Technology, 2013. (in Chinese)

Full-Text