针对中国西部地区辐射资源充沛但观测资料匮乏的特点,提出一种基于辐照度观测数据、遥感数据、 McClear和随机森林算法的太阳辐照度超短期预测方法,并重点分析遥感数据对辐照度预测效果的影响。结果表明：添加遥感数据能够优化不同时间步长的辐照度预测效果,并能显著降低平均绝对百分比误差（MAPE）值高于40%的预测大误差出现概率。同时,添加遥感数据对预测效果的提升随时间步长呈线性增加关系,nRMSE的差值变化范围从2.08%变为13.81%;nMAE的差值从1.64%变化为14.52%;R2的差值随时间步长的变化最为明显,从-0.03变为-0.43。但值得注意的是,添加卫星数据会显著增加模型的建立和超参寻优时间。
Abstract
In view of the characteristics of abundant radiation resources but lack of observation data in China, this study proposes a short-term solar irradiance forecasting method based on radiation observation data, remote sensing data, McClear, and random forest algorithm, and focuses on analyzing the impact of remote sensing data on radiation forecasting effectiveness. The results show that adding remote sensing data can optimize the forecasting effectiveness at different time horizons and significantly reduce the probability of large prediction errors with a mean absolute percentage error （MAPE） value exceeding 40%. Additionally, the improvement of the forecasting effectiveness with the addition of remote sensing data increases linearly with the time horizon. The difference range of normalized root mean square error （nRMSE） changes from 2.08% to 13.81%, the difference of normalized mean absolute error （nMAE） changes from 1.64% to 14.52%, the difference of R2 shows the most significant change with the time step, changing from -0.03 to -0.43. However, it is worth noting that adding satellite data will significantly increase the time required for model establishment and hyperparameter optimization.
关键词
太阳辐照度 /
预测 /
机器学习 /
FY-4A /
晴空模型
Key words
solar irradiance /
forecasting /
machine learning /
FY-4A /
clear sky model
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] YANG D E, KLEISSL J, GUEYMARD C A, et al.History and trends in solar irradiance and PV power forecasting: a preliminary assessment and review using text mining[J].Solar energy, 2018, 168: 60-101.
[2] JIA D Y,YANG L N, LV T, et al.Evaluation of machine learning models for predicting daily global and diffuse solar radiation under different weather/pollution conditions[J]. Renewable energy, 2022, 187: 896-906.
[3] 胡斯勒图, 施建成, 李明, 等. 基于卫星数据的地表下行短波辐射估算: 方法、 进展及问题[J]. 中国科学: 地球科学, 2020, 50(7): 887-902.
HU S L T, SHI J C,LI M, et al. A review of the estimation of downward surface shortwave radiation based on satellite data:methods, progress and problems[J]. Scientia sinica eerrae, 2020, 50(7): 887-902.
[4] 蒋俊霞, 高晓清. 光伏系统气候效应及影响机理研究进展[J]. 高原气象, 2022, 41(4): 953-962.
JIANG J X, GAO X Q.Research progress on climate effect and influence mechanism of photovoltaic systems[J]. Plateau meteorology, 2022, 41(4): 953-962.
[5] MANUEL Á J, GABRIEL R J, ANTONIO O S, et al.Hybrid techniques to predict solar radiation using support vector machine and search optimization algorithms:a review[J]. Applied sciences, 2021, 11(3): 1044.
[6] ZHANG L, MA W, ZHANG D.Stacked sparse autoencoder in PolSAR data classification using local spatial information[J]. IEEE geoscience and remote sensing letters, 2016, 13(9): 1359-1363.
[7] HINTON G E, SALAKHUTDINOV R R.Reducing the dimensionality of data with neural networks[J]. Science,2006, 313(5786): 504-507.
[8] YAGLI G M, YANG D E, SRINIVASAN D.Automatic hourly solar forecasting using machine learning models[J]. Renewable and sustainable energy reviews, 2019, 105: 487-498.
[9] MEGIA F A, KURTZ B, LEVIS A, et al.Cloud tomography applied to sky images: a virtual testbed[J].Solar energy, 2018, 176: 287-300.
[10] KUHN P, WILBERT S, PRAHL C, et al.Shadow camera system for the generation of solar irradiance maps[J]. Solar energy, 2017, 157: 157-170.
[11] ANDRÉ M, PEREZ R, SOUBDHAN T, et al.Preliminary assessment of two spatio-temporal forecasting technics for hourly satellite-derived irradiance in a complex meteorological context[J]. Solar energy, 2019, 177: 703-712.
[12] HARTY T M, HOLMGREN W F, LORENZO A T, et al.Intra-hour cloud index forecasting with data assimilation[J]. Solar energy, 2019, 185: 270-282.
[13] WU E, CLEMESHA R E S, KLEISSL J. Coastal Stratocumulus cloud edge forecasts[J]. Solar energy, 2018, 164: 355-369.
[14] MARION B, SIMITH B.Photovoltaic system derived data for determining the solar resource and for modeling the performance of other photovoltaic systems[J]. Solar energy, 2017, 147: 349-357.
[15] KISI O, PARMAR K S.Application of least square support vector machine and multivariate adaptive regression spline models in long term prediction of river water pollution[J].Journal of hydrology, 2016, 534: 104-112.
[16] DEO R C,WEN X H, Q F I. A wavelet-coupled support vector machine model for forecasting global incident solar radiation using limited meteorological dataset[J]. Applied energy, 2016, 168: 568-593.
[17] YAGLI G M, YANG D Z, GANDHI O, et al.Can we justify producing univariate machine-learning forecasts with satellite-derived solar irradiance?[J]. Applied energy,2020, 259: 114122.
[18] HUANG C L,SHI H R, GAO L, et al.Fengyun-4 geostationary satellite-based solar energy nowcasting system and its application in North China[J]. Advances in atmospheric sciences, 2022, 39(8): 1316-1328.
[19] YANG D Z.Choice of clear-sky model in solar forecasting[J]. Journal of renewable and sustainable energy, 2020, 12(2): 026101.
[20] JIA D Y, HUA J J, WANG L W, et al.Estimations of global horizontal irradiance and direct normal irradiance by using fengyun-4A satellite data in Northern China[J].Remote sensing, 2021, 13(4): 790.
[21] BREIMAN L.Bagging predictors[J]. Machine learning,1996, 24(2): 123-140.
[22] FENG Y,GONG D Z, ZHANG Q W, et al.Evaluation of temperature-based machine learning and empirical models for predicting daily global solar radiation[J]. Energy conversion and management, 2019, 198: 111780.
[23] YANG D Z.Making reference solar forecasts with climatology, persistence, and their optimal convex combination[J]. Solar energy, 2019, 193: 981-985.
[24] YANG D Z.A universal benchmarking method for probabilistic solar irradiance forecasting[J]. Solar energy,2019, 184: 410-416.
基金
国家自然科学基金（42305128）; 甘肃省教育厅：高校教师创新基金项目（2023B-151）; 甘肃省科技计划项目（21JR7RA546）
PDF(9815 KB)
