融合时空特征的光伏气象因子预测模型

doi:10.3778/j.issn.1002-8331.2105-0440

摘要/Abstract

摘要： 光伏发电极易受到天气的影响而具有波动性和不确定性，因此对气象因子的准确预测对光伏电站的运维具有重要意义。提出了一种基于深度学习的时空特征融合模型，实现对光伏气象因子的精准预测。在时间维度上，设计了一种改进的长短期记忆模块，融合注意力机制和遗传算法，得到最优注意力参数以提高预测精度；在空间维度上，将光伏电站所在区域按照经纬度划分，利用张量分解对区域内气象因子进行预测。在中国东南部某光伏系统的真实数据集上，对该模型的有效性进行了评估。结果表明，该模型在时间维度和空间维度均具有较高预测精度，同时对稀疏数据有较强的鲁棒性。

关键词: 气象因子预测, 长短期记忆网络, 张量分解, 注意力机制, 遗传算法

Abstract: Photovoltaic power generation is volatile and uncertain due to its high susceptibility to weather, so accurate prediction of meteorological factors is important for the operation and maintenance of PV plants. A deep learning based spatial-temporal feature fusion prediction model is proposed to achieve accurate prediction of PV weather factors. In the temporal dimension, an improved long short-term memory network module is designed to integrate the attention mechanism and genetic algorithm to obtain the optimal attention parameters to improve the prediction accuracy; in the spatial dimension, the area where the PV plant is located is divided according to latitude and longitude, and the tensor decomposition is used to predict the meteorological factors in the area spatially. In the experiments, the effectiveness of the proposed model is evaluated using real datasets of PV plants and compared with the existing methods. The results show that the model has high prediction accuracy in both time and spatial dimensions, and has strong robustness to sparse data.

Key words: prediction of meteorological factors, long short-term memory network, tensor decomposition, attention mechanism, genetic algorithm

李金中, 王小明, 谢毓广, 高博, 汪勋婷. 融合时空特征的光伏气象因子预测模型[J]. 计算机工程与应用, 2022, 58(23): 285-292.

LI Jinzhong, WANG Xiaoming, XIE Yuguang, GAO Bo, WANG Xunting. Photovoltaic Meteorological Factor Prediction Model Fusing Spatial and Temporal Features[J]. Computer Engineering and Applications, 2022, 58(23): 285-292.

参考文献

[1] SHI J，LEE W J，LIU Y，et al.Forecasting power output of photovoltaic systems based on weather classification and support vector machines[J].IEEE Transactions on Industry Applications，2012，48（3）：1064-1069.
[2] KOLDA T G，BADER B W.Tensor decompositions and applications[J].SIAM Review，2009，51（3）：455-500.
[3] LAW E W，PRASAD A A，KAY M，et al.Direct normal irradiance forecasting and its application to concentrated solar thermal output forecasting—a review[J].Solar Energy，2014，108：287-307.
[4] LORENC A.Analysis methods for numerical weather prediction[J].Quarterly Journal of the Royal Meteorological Society，1986，112（474）：1177-1194.
[5] WANG F，MI Z，SU S，et al.Short-term solar irradiance forecasting model based on artificial neural network using statistical feature parameters[J].Energies，2012，5（5）：1355-1370.
[6] MILLER S D，ROGERS M A，HAYNES J M，et al.Short-term solar irradiance forecasting via satellite/model coupling[J].Solar Energy，2018，168：102-117.
[7] RYU A，ITO M，ISHII H，et al.Preliminary analysis of short-term solar irradiance forecasting by using total-sky imager and convolutional neural network[C]//2019 IEEE PES GTD Grand International Conference and Exposition Asia（GTD Asia），Bangkok，Thailand，March 20-23，2019.New York，USA：IEEE，2019：627-631.
[8] MCCANDLESS T，JIMéNEZ P A.Examining the potential of a random forest derived cloud mask from GOES-R satellites to improve solar irradiance forecasting[J].Energies，2020，13（7）：1671.
[9] 蒋俊霞，高晓清，吕清泉，等.基于地基云图的云跟踪与太阳辐照度超短期预报方法研究[J].太阳能学报，2020，41（5）：351-358.
JIANG J X，GAO X Q，LYU Q Q，et al.Study on cloud tracking and solar irradiance ultra-short-term forecasting based on TSI images[J].Acta Energiae Solaris Sinica，2020，41（5）：351-358.
[10] HOCAOGLU F O，SERTTAS F.A novel hybrid（Mycielski-Markov） model for hourly solar radiation forecasting[J].Renewable Energy，2017，108：635-643.
[11] HUANG X，SHI J，GAO B，et al.Forecasting hourly solar irradiance using hybrid wavelet transformation and elman model in smart grid[J].IEEE Access，2019，7：139909-139923.
[12] DEWANGAN C L，SINGH S N，CHAKRABARTI S.Solar irradiance forecasting using wavelet neural network[C]//2017 IEEE PES Asia-Pacific Power and Energy Engineering Conference（APPEEC），Bangalore，India，Nov 8-11，2017.New York，USA：IEEE，2017：1-6.
[13] 田翠霞，黄敏，朱启兵.基于EMD-LMD-LSSVM联合模型的逐时太阳辐照度预测[J].太阳能学报，2018，39（2）：504-512.
TIAN C X，HUANG M，ZHU Q B.Hourly solar irradiance forecast based on EMD-LMD-LSSVM joint model[J].Acta Energiae Solaris Sinica，2018，39（2）：504-512.
[14] THEOCHARIDES S，KYNIGOS M，THERISTIS M，et al.Intra-day solar irradiance forecasting based on artificial neural networks[C]//2019 IEEE 46th Photovoltaic Specialists Conference（PVSC），Chicago，USA，June 16-21，2019.New York，USA：IEEE，2019：1628-1631.
[15] SIVANEASAN B，YU C Y，GOH K P.Solar forecasting using ANN with fuzzy logic pre-processing[J].Energy Procedia，2017，143：727-732.
[16] ZOU L，LIN A W，WANG L C，et al.Monthly mean global solar radiation modeling using artificial neural network technique in southeast hill areas，China during 1993—2013[C]//Proceedings of 2015 International Conference on Energy and Mechanical Engineering，Wuhan，China，2016：394-401.
[17] 曹英丽，方诗琦，王洋，等.分布式光伏发电特性与气象影响因子诊断分析[J].沈阳农业大学学报，2018，49（3）：363-370.
CAO Y L，FANG S Q，WANG Y，et al.Diagnostic analysis of distributed photovoltaic power characteristics and the impact of meteorological factors[J]，Journal of Shenyang Agricultural University，2018，49（3）：363-370.
[18] WEYB J A，DURRAN D R，CARUANA R.Improving data‐driven global weather prediction using deep convolutional neural networks on a cubed sphere[J].Journal of Advances in Modeling Earth Systems，2020，12（9）：e2020MS002109.
[19] 赵书强，尚煜东，杨燕燕，等.基于长短期记忆神经网络的地表太阳辐照度预测[J].太阳能学报，2021，42（3）：383-388.
ZHAO S Q，SHANG Y D，YANG Y Y，et al.Prediction of surface solar irradiance based on long and short-term memory neural network[J].Acta Energiae Solaris Sinica，2021，42（3）：383-388.
[20] ZHENG Y，LIU T，WANG Y，et al.Diagnosing New York city’s noises with ubiquitous data[C]//Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing，Seattle，USA，Sep 13-17，2014.New York：ACM，2014：715-725.
[21] WHITLEY D.A genetic algorithm tutorial[J].Statistics and Computing，1994，4（2）：65-85.
[22] KARATZOGLOU A，AMATRIAIN X，BALTRUNAS L，et al.Multiverse recommendation：n-dimensional tensor factorization for context-aware collaborative filtering[C]//Proceedings of the Fourth ACM Conference on Recommender Systems，Barcelona，Spain，Sep 26-30，2010.New York：ACM，2010：79-86.
[23] AHYAR L F，SUVANTO S，ARIFIANTO A.Firefly algorithm-based hyperparameters setting of DRNN for weather prediction[C]//2020 International Conference on Data Science and Its Applications（ICoDSA），Aug 5-6，2020.New York：IEEE，2020：1-5.