检验评估2022年CMA-WSP、CMA-MESO、NCEP-GFS这3种数值预报模式在河北省30个光伏电站的总辐射辐照度（GHI）预报效果。结果表明：CMA-WSP预报误差最低,相关系数最高,CMA-MESO次之;CMA-WSP预报系统性偏高,CMA-MESO和NCEP-GFS预报系统性偏低。CMA-MESO在偏差200 W/m2以上的大误差比例最低、CMA-WSP次之,各数值模式均有偏差在700 W/m2以上的样本。在雾、霾、沙尘天气过程中,CMA-WSP的预报效果最好,在降雨、降雪、连阴雨、强对流天气过程中,CMA-MESO的预报效果最好。对于不同位置的光伏电站,3种数值模式在不同天气过程中的预报效果有明显差异,有些电站在各数值模式对各天气过程的预报中整体误差都较小,有些电站则整体误差较大。
Abstract
This study evaluates the performance of 3 numerical prediction models： CMA-WSP, CMA-MESO and NCEP-GFS, of the global horizontal irradiance （GHI） at 30 PV stations in Hebei Province in 2022. The results show that the CMA-WSP has the lowest forecasting error and the highest correlation coefficient, followed by CMA-MESO. The CMA-WSP exhibits a systematic overestimation, while both CMA-MESO and NCEP-GFS show a systematic underestimation. The CMA-MESO has the lowest large error ratio （with a deviation of over 200 W/m2）, while the CMA-WSP coming second. All models show samples with deviation of more than 700 W/m2. During fog, haze, and sandstorm weather processes, the CMA-WSP performs the best, while the CMA-MESO performs the best during rainfall, snowfall, continuous rainy periods, and severe convective weather processes. For PV stations at different locations, the forecasting performance of the three models varies significantly across different weather processes. For some stations, the overall forecasting errors of all numerical models are relatively small across different weather processes, while for others, the overall errors are larger.
关键词
太阳辐射 /
预报 /
数值模式 /
高影响天气 /
光伏电站 /
检验
Key words
solar radiation /
forecasting /
numerical model /
high-impact weather /
PV station /
verification
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 国家能源局2025年全国电力统计数据[EB/OL]. (2026-01-29). https://www.nea.gov.cn/20260129/6874f211acd0417eab7ac10c3061a7c2/c.html.
NEA national electric power industry statistical data of 2025[EB/OL]. (2026-01-29). https://www.nea.gov.cn/20260129/6874f211acd0417eab7ac10c3061a7c2/c.html.
[2] GB/T 40607—2021, 调度侧风电或光伏功率预测系统技术要求[S].
GB/T 40607—2021, Technical requirements for dispatching side forecasting system of wind or photovoltaic power[S].
[3] 贾东于, 李开明, 高晓清, 等. 基于FY-4A和机器学习的太阳辐照度超短期预测[J]. 太阳能学报, 2024, 45(4): 578-583.
JIA D Y, LI K M, GAO X Q, et al.Nowcasting prediction of solar irradiance based on FY-4A and machine learning[J]. Acta energiae solaris sinica, 2024, 45(4): 578-583.
[4] KRISHNAN N, KUMAR K R, INDA C S.How solar radiation forecasting impacts the utilization of solar energy: a critical review[J]. Journal of cleaner production, 2023, 388: 135860.
[5] DOU W J, WANG K, SHAN S, et al.Day-ahead numerical weather prediction solar irradiance correction using a clustering method based on weather conditions[J]. Applied energy, 2024, 365: 123239.
[6] INMAN R H, PEDRO H T C, COIMBRA C F M. Solar forecasting methods for renewable energy integration[J]. Progress in energy and combustion science, 2013, 39(6): 535-576.
[7] LARA-FANEGO V, RUIZ-ARIAS J A, POZO-VÁZQUEZ D, et al. Evaluation of the WRF model solar irradiance forecasts in Andalusia (southern Spain)[J]. Solar energy, 2012, 86(8): 2200-2217.
[8] MATHIESEN P, KLEISSL J.Evaluation of numerical weather prediction for intra-day solar forecasting in the continental United States[J]. Solar energy, 2011, 85(5): 967-977.
[9] SAWADOGO W, FERSCH B, BLIEFERNICHT J, et al.Evaluation of the WRF-solar model for 72-hour ahead forecasts of global horizontal irradiance in West Africa: a case study for Ghana[J]. Solar energy, 2024, 271: 112413.
[10] 袁彬, 于廷照, 申彦波, 等. 基于人工智能多模式集成的光伏电站总辐射预报方法研究[J]. 太阳能学报, 2025, 46(4): 291-300.
YUAN B, YU T Z, SHEN Y B, et al.Research on artificial intelligence-based multi-model ensemble forecast of global horizontal irradiance in PV station[J]. Acta energiae solaris sinica, 2025, 46(4): 291-300.
[11] YUAN B, SHEN Y B, DENG H, et al.Multimodel ensemble forecast of global horizontal irradiance at PV power stations based on dynamic variable weight[J]. Journal of tropical meteorology, 2024, 30(3): 327-336.
[12] LEE J, WANG W, HARROU F, et al.Reliable solar irradiance prediction using ensemble learning-based models: a comparative study[J]. Energy conversion and management, 2020, 208: 112582.
[13] 李芬, 刘楚琦, 于淏, 等. 基于天气分型的水平面总辐射及散射辐射建模分析[J]. 太阳能学报, 2024, 45(4): 560-568.
LI F, LIU C Q, YU H, et al.Modeling and analysis of horizontal plane global solar radiation and diffuse solar radiation based on weather type classification[J]. Acta energiae solaris sinica, 2024, 45(4): 560-568.
[14] TZOUMANIKAS P, NIKITIDOU E, BAIS A F, et al.The effect of clouds on surface solar irradiance, based on data from an all-sky imaging system[J]. Renewable energy, 2016, 95: 314-322.
[15] 秦放, 李登宣, 丁煌, 等. 基于CERES数据分析中国云对地表太阳辐射影响特征[J]. 太阳能学报, 2022, 43(9): 8-14.
QIN F, LI D X, DING H, et al.Chracteristics of surface solar radiation and cloud fraction in china based on CERES[J]. Acta energiae solaris sinica, 2022, 43(9): 8-14.
[16] 李张群, 肖子牛, 丁煌, 等. 河南地区降水量和云量与地面太阳短波辐射的关系[J]. 气候与环境研究, 2022, 27(4): 504-512.
LI Z Q, XIAO Z N, DING H, et al.Relationship of surface solar shortwave radiation with precipitation and cloud cover in Henan Province[J]. Climatic and environmental research, 2022, 27(4): 504-512.
[17] HIRATA K, TAKEDA Y, GOTO S, et al.Forecasting solar irradiance during transitional weather conditions using ensemble modeling[J]. Energy procedia, 2015, 69: 1957-1966.
[18] DOU W J, WANG K, SHAN S, et al.Evaluation of performance for day-ahead solar irradiance forecast using numerical weather prediction[J]. Journal of renewable and sustainable energy, 2024, 16(4): 043703.
[19] 李芬, 李春阳, 闫全全, 等. 多时间尺度下光伏出力爬坡事件概率建模与评估研究[J]. 太阳能学报, 2019, 40(11): 3289-3298.
LI F, LI C Y, YAN Q Q, et al.PV power ramp events probability modeling and assessment in multiple time scales[J]. Acta energiae solaris sinica, 2019, 40(11): 3289-3298.
[20] BALSAMO G, RABIER F, BALMASEDA M, et al.Recent progress and outlook for the ECMWF integrated forecasting system[C]//EGU23, the 25th EGU General Assembly. Vienna, Austria, 2023.
[21] 陈红梅, 张立波, 娄伟平. 近50a华北平原日照时数的时空特征及其影响因素[J]. 气象科学, 2012, 32(5): 573-579.
CHEN H M, ZHANG L B, LOU W P.Temporal and spatial variations of sunshine duration and its influence factors in the North China plain in recent 50 years[J]. Journal of the meteorological sciences, 2012, 32(5): 573-579.
[22] GB/T 37526—2019, 太阳能资源评估方法[S].
GB/T 37526—2019, Assessment method for solar energy resource[S].
[23] PARK S K.Numerical weather prediction: East Asian perspectives[M]. Cham: Springer, 2023: 63-104.
[24] 黄丽萍, 邓莲堂, 王瑞春, 等. CMA-MESO关键技术集成及应用[J]. 应用气象学报, 2022, 33(6): 641-654.
HUANG L P, DENG L T, WANG R C, et al.Key technologies of CMA-MESO and application to operational forecast[J]. Journal of applied meteorological science, 2022, 33(6): 641-654.
基金
国家重点研发计划（2024YFF0809204）; 新疆科技厅上海合作组织科技伙伴计划及国际科技合作计划项目（2023E01011）; 中国气象局揭榜挂帅科技项目（CMAJBGS202209b）
PDF(5046 KB)
