Please wait a minute...
 
Remote Sensing for Natural Resources    2022, Vol. 34 Issue (2) : 105-111     DOI: 10.6046/zrzyyg.2021151
|
Revision of solar radiation product ERA5 based on random forest algorithm
WANG Xuejie1(), SHI Guoping2(), ZHOU Ziqin1, ZHEN Yang1
1. Changwang School of Honors, Nanjing University of Information Science and Technology, Nanjing 210044, China
2. School of Geographical Sciences, Nanjing University of Information Science and Technology, Nanjing 210044, China
Download: PDF(3702 KB)   HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks    
Abstract  

This study performed a multi-scale error analysis of the mean surface downward shortwave radiation flux product ERA5 (0.25° × 0.25°) of the European Centre for Medium-Range Weather Forecasts (ECMWF) using 93 pieces of solar radiation hourly data in 2013 of China. Subsequently, this study revised and analyzed the total radiation product ERA5 by training the random forest model using various relevant elements such as meteorological and geographic ones. Finally, the model was used to obtain the map of revised hourly radiation spatial distribution. As a result, the reanalyzed data can be better applied in industries such as agriculture, electric power, and urban construction. The results are as follows. ① The MAE, RMSE, and R values between the ERA5 solar radiation and the measured values of stations in 2013 were 27.60 W/m2, 29.87 W/m2, and 0.97 respectively. Moreover, the ERA5 values were higher than the measured values of stations. ② The accuracy was improved after the revision using the random forest model. After revision, the MAE, RMSE, and R values between the ERA5 solar radiation and the measured values of stations were 3.34 W/m2, 3.85 W/m2, and 1.00, respectively, indicating that correlation was significantly improved. ③ The spatial macroscopic distribution patterns of radiation before and after revision were consistent, but the ERA5 radiation value significantly decreased in local areas.
Keywords ERA5 products      solar radiation      random forest revision     
ZTFLH:  P422.1  
Corresponding Authors: SHI Guoping     E-mail: 201883330052@nuist.edu.cn;shiguopingnj@163.com
Issue Date: 20 June 2022
Service
E-mail this article
E-mail Alert
RSS
Articles by authors
Xuejie WANG
Guoping SHI
Ziqin ZHOU
Yang ZHEN
Cite this article:   
Xuejie WANG,Guoping SHI,Ziqin ZHOU, et al. Revision of solar radiation product ERA5 based on random forest algorithm[J]. Remote Sensing for Natural Resources, 2022, 34(2): 105-111.
URL:  
https://www.gtzyyg.com/EN/10.6046/zrzyyg.2021151     OR     https://www.gtzyyg.com/EN/Y2022/V34/I2/105
模型参数 参数取值
子树数量 56
衡量分裂质量的性能 GI
最佳分裂点时考虑的属性数目 auto
树的最大深度 30
叶节点最小样本数 5
分割内部节点的最小样本数 8
Bootstrap抽样 TRUE
是否使用袋外样本 FALSE
随机数生成器使用的种子 None
Tab.1  Value of model parameters
次数 决定系数R2
1 0.865
2 0.851
3 0.854
4 0.868
5 0.837
平均值 0.855
Tab.2  Result of model 5-fold cross-validation
Fig.1  Comparison of feature importance
Fig.2  Scatter diagram of hourly radiation before and after revision in four months
Fig.3  Monthly radiation mean value before and after revision
误差指标 订正前 订正后
MAE/(W·m-2) 27.60 3.34
RMSE/(W·m-2) 29.87 3.85
R 0.97 1.00
Tab.3  Comparison of three error indexes
Fig.4  Comparison of AE,MAE,RMSE and R before and after revision
站号 站点 纬度/(°) 经度/(°)
51628 阿克苏 N41.17 E80.23
52754 刚察 N37.33 E100.13
54161 长春 N43.90 E125.22
56196 绵阳 N31.45 E104.75
58737 建瓯 N27.05 E118.32
59644 北海 N21.45 E109.13
Tab.4  Information of six stations
站点 订正前 订正后 MAE
改善率/%		 RMSE
改善率/%		 R提高量
MAE RMSE R MAE RMSE R
阿克苏 237.83 288.28 0.50 62.49 96.82 0.94 73.72 66.42 0.44
刚察 170.13 208.48 0.75 74.80 117.00 0.92 56.03 43.88 0.17
长春 65.40 97.32 0.93 59.50 96.12 0.93 9.01 1.23 0
绵阳 109.09 148.87 0.83 74.47 113.13 0.89 31.74 24.01 0.06
建瓯 74.66 120.87 0.90 72.25 112.71 0.92 3.23 6.75 0.02
北海 121.11 165.75 0.84 85.21 129.37 0.89 29.64 21.95 0.05
Tab.5  Analysis of error indexes in six stations before and after revision
Fig.5-1  Spatial distribution of solar radiation before and after revision
Fig.5-2  Spatial distribution of solar radiation before and after revision
[1] 刘钊, 于学峰. 太阳驱动地球环境变化研究进展[J]. 自然杂志, 2012, 34(3):154-156,160.
[1] Liu Z, Yu X F. The progress in study of the effects of the solar variation on the earth environment[J]. Chinese Journal of Nature, 2012, 34(3):154-156,160.
[2] Roberto R, Renzo R. Distributed estimation of incoming direct solar radiation over a drainage basin[J]. Journal of Hydrology, 1995, 166(3):461-478.
doi: 10.1016/0022-1694(94)05099-J url: https://linkinghub.elsevier.com/retrieve/pii/002216949405099J
[3] 赵康, 桂雪晨, 葛坚. 高大空间中太阳辐射对热舒适的影响及室内参数设计[J]. 太阳能学报, 2019, 40(9):2655-2662.
[3] Zhao K, Gui X C, Ge J. Influence of solar radiation on thermal comfort in large spaces and corresponding design of indoor parameters[J]. Acta Energiae Solaris Sinica, 2019, 40(9):2655-2662.
[4] 张欣欣, 景丽. 太阳能热水系统在建筑给排水设计中的应用[J]. 智能建筑与智慧城市, 2021(2):88-89.
[4] Zhang X X, Jing L. Application of solar water heating system in the design of building water supply and drainage[J]. Intelligent Building and City Information, 2021(2):88-89.
[5] 张佳飞. 多时间尺度太阳辐射估算模型[D]. 重庆: 西南大学, 2013.
[5] Zhang J F. Estimation models of solar radiation at different time scales[D]. Chongqing: Xi’nan University, 2013.
[6] 吕宁. 近年来中国地表太阳辐射时空变化及影响因素分析[D]. 北京: 中国科学院地理科学与资源研究所, 2009.
[6] Lyu N. Analysis of spatio-temporal variation of surface downward shortwave radiation and associated effecting factors over China in recent years[D]. Beijing: Institute of Geographic Sciences and Natural Resources Research, CAS, 2009.
[7] 蔡子颖. 我国华东、中南地区地面太阳辐射变化规律及其原因分析[D]. 南京: 南京信息工程大学, 2009.
[7] Cai Z Y. Analysis on the variation of surface solar radiation in East China,Central South China and its causes[D]. Nanjing: Nanjing University of Information Science and Technology, 2009.
[8] 张星星, 吕宁, 姚凌, 等. ECMWF地表太阳辐射数据在我国的误差及成因分析[J]. 地球信息科学学报, 2018, 20(2):254-267.
doi: 10.12082/dqxxkx.2018.170381
[8] Zhang X X, Lyu N, Yao L, et al. Error analysis of ECMWF surface solar radiation data in China[J]. Journal of Geo-Information Science, 2018, 20(2):254-267.
[9] Zhang X T, Liang S L, Wang G X, et al. Evaluation of the reanalysis surface incident shortwave radiation products from NCEP,ECMWF,GSFC,and JMA using satellite and surface observations[J]. Remote Sensing, 2016, 8(3):225-237.
doi: 10.3390/rs8030225 url: http://www.mdpi.com/2072-4292/8/3/225
[10] Wang A H, Zeng X B. Evaluation of multireanalysis products with in situ observations over the Tibetan Plateau[J]. Journal of Geophysical Research: Atmospheres, 2012, 117:D05102.
[11] 陈昱文, 黄小猛, 李熠, 等. 基于ECMWF产品的站点气温预报集成学习误差订正[J]. 应用气象学报, 2020, 31(4):494-503.
[11] Chen Y W, Huang X M, Li Y, et al. Ensemble learning for bias correction of station temperature forecast based on ECMWF products[J]. Journal of Applied Meteorological Science, 2020, 31(4):494-503.
[12] 李净, 温松楠. 基于3种机器学习法的太阳辐射模拟研究[J]. 遥感技术与应用, 2020, 35(3):615-622.
[12] Li J, Wen S N. Simulation of solar radiation based on three machine learning methods[J]. Remote Sensing Technology and Application, 2020, 35(3):615-622.
[13] Benali L, Notton G, Fouilloy A, et al. Solar radiation forecasting using artificial neural network and random forest methods: Application to normal beam,horizontal diffuse and global components[J]. Renewable Energy, 2019, 132:871-884.
doi: 10.1016/j.renene.2018.08.044
[14] Yu W, Zhang X T, Hou N, et al. Estimation of surface downward shortwave radiation over China from AVHRR data based on four machine learning methods[J]. Solar Energy, 2019, 177:32-46.
doi: 10.1016/j.solener.2018.11.008 url: https://linkinghub.elsevier.com/retrieve/pii/S0038092X18311083
[15] 许立兵, 王安喜, 汪纯阳, 等. 基于机器学习的海洋环境预报订正方法研究[J]. 海洋通报, 2020, 39(6):695-704.
[15] Xu L B, Wang A X, Wang C Y, et al. Research on correction method of marine environment prediction based on machine learning[J]. Marine Science Bulletin, 2020, 39(6):695-704.
[16] 陈有龙, 宁雨珂, 唐荣年, 等. 基于时空独立的随机森林模型对海南热带气温数值预报的订正[J]. 海南大学学报(自然科学版), 2020, 38(4):356-364.
[16] Chen Y L, Ning Y K, Tang R N, et al. Tropical temperature correction for numerical forecast in Hainan based on spatiotemporal independence random forest model[J]. Natural Science Journal of Hainan University, 2020, 38(4):356-364.
[17] 芦华, 谢旻, 吴钲, 等. 基于机器学习的成渝地区空气质量数值预报PM2.5订正方法研究[J]. 环境科学学报, 2020, 40(12):4419-4431.
[17] Lu H, Xie M, Wu Z, et al. Adjusting PM2.5 prediction of the numerical air quality forecast model based on machine learning methods in Chengyu region[J]. Acta Scientiae Circumstantiae, 2020, 40(12):4419-4431.
[18] Babar B, Luppino L T, BostrÖm T, et al. Random forest regression for improved mapping of solar irradiance at high latitudes[J]. Solar Energy, 2019, 198:81-92.
doi: 10.1016/j.solener.2020.01.034 url: https://linkinghub.elsevier.com/retrieve/pii/S0038092X20300426
[1] WEI Shenglong, CHEN Zhibiao, CHEN Zhiqiang, WANG Qiuyun, MA Xiuli, YAN Xinyu. Simulation of the total solar radiation over micro-landform and correlation between the solar radiation and the land surface temperature[J]. REMOTE SENSING FOR LAND & RESOURCES, 2017, 29(1): 129-135.
[2] Wan Yuqing, Wu Junhu, Lei Xuewu . FACTORS NEED TO BE THOUGHT ABOUT WHILE USING REMOTE SENSING IMAGES TO EXTRACT COAL FIRE INFORMATION[J]. REMOTE SENSING FOR LAND & RESOURCES, 1998, 10(1): 33-39.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
京ICP备05055290号-2
Copyright © 2017 Remote Sensing for Natural Resources
Support by Beijing Magtech