针对辐射周期趋势及外部影响特征捕获不足的问题,提出一种线性分解和周期增强Informer的地表太阳辐射短临预报方法。首先,改进灰色关联度方法,获取历史辐射与多种外部气象因素关联度,提取16种高相关外部气象特征建立高关联特征集,强化捕捉辐射与气象因素之间的复杂关系的能力;其次,在基于Transformer解决方案的基础上引入周期性嵌入层和ReLU激活函数,为模型提供更准确、合理的周期时间特征和辐射变化区间。最后,在Informer后增加平滑序列分解线性层,将Autoformer中的分解方案和FEDformer中的线性层相结合,进一步增强捕捉时序数据中周期性和季节性成分的能力。实验结果表明：该IDL方法结合外部气象特征能极好地提高模型短临预报效果,精度高于近年来基于Transformer系列的解决方案;比DLinear均方误差最高减少30.6%。
Abstract
In response to the inadequacy in capturing trend and periodic features, a method for short arrival prediction of surface solar radiation :s proposed decomposition and periodic-enhancement linear Informer. Firstly, an improved grey relational analysis method captures the correlation between historical irradiance and various external meteorological factors. This method extracts 16 highly correlated external meteorological features, thus enhancing the model's ability to capture the intricate relationship between irradiance and meteorological factors. Subsequently, the Informer model is augmented with periodic embedding layers and ReLU activation functions, which better represents the periodic variations in solar irradiance and provides more accurate temporal features. Finally, an integration of the decomposition scheme from Autoformer and the linear layer from FEDformer as a decomposition linear layer after the Informer. This amalgamation enhances the model’s capability to capture the periodic trends and seasonal components in time series solar irradiance data. Experimental results demonstrate that the proposed Informer Decomposition Linear model, in conjunction with external meteorological features, remarkably improves short-term forecasting performance, surpassing the accuracy of recently prominent Transformer-based approaches. In comparison to the best-performing DLinear model, the maximum
关键词
太阳辐射 /
Informer /
Transformer /
平滑序列线性分解 /
周期嵌入 /
灰色关联度
Key words
solar irradiance /
Informer /
Transformer /
smoothing sequential linear decomposition /
periodic embedding /
grey relational degree
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] HUANG G H, LI Z Q, LI X, et al.Estimating surface solar irradiance from satellites: past, present, and future perspectives[J]. Remote sensing of environment, 2019, 233: 111371.
[2] LI Q Q, ZHANG D P, YAN K.A solar irradiance forecasting framework based on the CEE-WGAN-LSTM model[J]. Sensors, 2023, 23(5): 2799.
[3] 沈钟平, 张华. 影响地面太阳辐射及其谱分布的因子分析[J]. 太阳能学报, 2009, 30(10): 1209-1215.
SHEN Z P, ZHANG H.Analysis on the factors affecting surface solar radiation and its spectral distribution[J]. Acta energiae solaris sinica, 2009, 30(10): 1209-1215.
[4] 郝玉珠, 李兴华, 胡亚男, 等. 内蒙古57年太阳能资源变化规律及影响因子[J]. 太阳能学报, 2021, 42(9): 145-151.
HAO Y Z, LI X H, HU Y N, et al.Change rules and influencing factors of 57 years of solar energy resources in Inner Mongolia[J]. Acta energiae solaris sinica, 2021, 42(9): 145-151.
[5] 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.
[6] 杜智涛, 姜明波, 杜晓勇, 等. 机器学习在气象领域的应用现状与展望[J]. 气象科技, 2021, 49(6): 930-941.
DU Z T, JIANG M B, DU X Y, et al.Application status and prospect of machine learning in meteorology[J]. Meteorological science and technology, 2021, 49(6): 930-941.
[7] AL-QANESS M A A, EWEES A A, FAN H, et al. Wind power prediction using random vector functional link network with capuchin search algorithm[J]. Ain shams engineering journal, 2023, 14(9): 102095.
[8] 普智勇, 夏攀, 张璐, 等. 机器学习与统计方法在太阳能预报中的比较性分析[J]. 太阳能学报, 2023, 44(7): 162-167.
PU Z Y, XIA P, ZHANG L, et al.Comparative analysis of machine learning and statistical methods in solar energy prediction[J]. Acta energiae solaris sinica, 2023, 44(7): 162-167.
[9] PATEL D, PATEL S, PATEL P, et al.Solar radiation and solar energy estimation using ANN and fuzzy logic concept: a comprehensive and systematic study[J]. Environmental science and pollution research, 2022, 29(22): 32428-32442.
[10] 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.
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.
[11] 王雪洁, 施国萍, 周子钦, 等. 基于随机森林算法对ERA5太阳辐射产品的订正[J]. 自然资源遥感, 2022, 34(2): 105-111.
WANG X J, SHI G P, ZHOU Z Q, et al.Revision of solar radiation product ERA5 based on random forest algorithm[J]. Remote sensing for natural resources, 2022, 34(2): 105-111.
[12] SALEHI A W, KHAN S, GUPTA G, et al.A study of CNN and transfer learning in medical imaging: advantages, challenges, future scope[J]. Sustainability, 2023, 15(7): 5930.
[13] ZHOU Y, LI Y Z, WANG D J, et al.A multi-step ahead global solar radiation prediction method using an attention-based transformer model with an interpretable mechanism[J]. International journal of hydrogen energy, 2023, 48(40): 15317-15330.
[14] FENG Y, HAO W P, LI H R, et al.Machine learning models to quantify and map daily global solar radiation and photovoltaic power[J]. Renewable and sustainable energy reviews, 2020, 118: 109393.
[15] VASWANI A, SHAZEER N, PARMAR N, et al.Attention is all you need[C]//Advances in Neural Information Processing Systems. Long Beach, USA: Curran Associates Inc., 2017: 5998-6008.
[16] BAI Y M, ZHAO J S.A novel transformer-based multi-variable multi-step prediction method for chemical process fault prognosis[J]. Process safety and environmental protection, 2023, 169: 937-947.
[17] POSPÍCHAL J, KUBOVČÍK M, DIRGOVÁ LUPTÁKOVÁ I. Solar irradiance forecasting with transformer model[J]. Applied sciences, 2022, 12(17): 8852.
[18] ZHOU T, MA Z, WEN Q, et al.FEDformer: frequency enhanced decomposed transformer for long-term series forecasting[C]//Proceedings of the 39th International Conference on Machine Learning. Baltimore USA, 2022.
[19] WEN Q S, ZHOU T, ZHANG C L, et al. Transformers in time series: a survey[EB/OL].2022: 2202.07125. https://arxiv.org/abs/2202.07125v5.
[20] ZENG A L, CHEN M X, ZHANG L, et al.Are transformers effective for time series forecasting?[J]. Proceedings of the AAAI conference on artificial intelligence, 2023, 37(9): 11121-11128.
[21] 王涛, 王旭, 许野, 等. 计及相似日的LSTM光伏出力预测模型研究[J]. 太阳能学报, 2023, 44(8): 316-323.
WANG T, WANG X, XU Y, et al.Study on LSTM photovoltaic output prediction model considering similar days[J]. Acta energiae solaris sinica, 2023, 44(8): 316-323.
[22] ZHOU H Y, ZHANG S H, PENG J Q, et al.Informer: beyond efficient transformer for long sequence time-series forecasting[J]. Proceedings of the AAAI conference on artificial intelligence, 2021, 35(12): 11106-11115.
[23] WU H, XU J, WANG J, et al.Autoformer: decomposition transformers with auto-correlation for long-term series forecasting[C]//Advances in Neural Information Processing Systems 34. Reed Hook, NY, USA: Curran Associates, Inc., 2021: 22419-22430.
[24] RAMPÁŠEK L, GALKIN M, DWIVEDI V P, et al. Recipe for a general, powerful, scalable graph transformer[C]//Advances in Neural Information Processing Systems 35. Reed Hook, NY, USA: Curran Associates, Inc., 2022: 14501-14515.
基金
高层次创新人才培养专项（B12402005）; 四川轻化工大学人才引进项目（2021RC16）; 教育部高等教育司产学合作协同育人项目（202101038016）
PDF(1515 KB)
