基于 LSTM 模型的光伏辐照度短期预报订正技术

曾荣阳 ,  雍佳 ,  武万里 ,  刘建宏 ,  龚晓丽

西北工程技术学报(中英文) ›› 2026, Vol. 25 ›› Issue (2) : 110 -119.

PDF (1548KB)
西北工程技术学报(中英文) ›› 2026, Vol. 25 ›› Issue (2) : 110 -119. DOI: 10.26974/j.cnki.XBGC.2026.02.002
新能源与新材料

基于 LSTM 模型的光伏辐照度短期预报订正技术

作者信息 +

LSTM-Based Model for Short-Term Correction of Photovoltaic Irradiance Forecasting

Author information +
文章历史 +
PDF (1585K)

摘要

为提高数值模式光伏辐照度短期预报的精度,利用中国气象局风能太阳能数值预报系统(CMA-WSP 2.0)中 2023-2025 年逐小时短期辐射预报产品及宁夏的光伏电站和气象观测站的地面观测辐照度资料,以模式预报辐射及相关气象要素为输入,构建融合长短期记忆(long short-term memory,LSTM)与注意力机制的序列订正模型(attention-enhanced LSTM sequence correction,ALSC)。其中,LSTM 用于刻画辐射时间序列的长期依赖关系,注意力机制用于自适应分配不同气象因子的权重,以未经订正的 CMA-WSP 2.0 预报产品作为对照;采用决定系数(R²)、均方根误差(RMSE)和平均绝对误差(MAE)等指标对模型订正效果进行检验。结果表明,使用 2025 年的资料为测试数据,订正后辐射预报的拟合优度 R²明显提高,RMSE 和 MAE 分别降低 10.57% 和 7.57%;以 3 月份数据分天空状况进行评估,阴天和多云条件下平均相对误差分别由 28.87% 和 25.47% 降至 3.06% 和 0.33%,误差改进率(Skill,表示订正后误差相对原始预报值误差的降低幅度)分别为 89.4% 和 98.7%,辐照度预报精度显著提升;在多云、辐射快速变化等天气条件下,模型对辐射突变的响应能力得到增强。研究可为光伏电站发电功率预测及电网调度优化提供客观参考和工程应用支撑。

Abstract

To improve the accuracy of short-term photovoltaic irradiance forecasts from numerical models, the study utilized the hourly short-term radiation forecast products from the China Meteorological Administration Wind and Solar Numerical Forecasting System (CMA-WSP 2.0) for the period 2023-2025, along with ground observational irradiance data from photovoltaic power stations and meteorological observation stations in Ningxia. A sequence correction model integrating a long short-term memory (LSTM) network with an Attention Mechanism (hereafter referred to as attention-enhanced LSTM sequence correction, ALSC) was constructed, in which the LSTM captured the long-term dependence of radiation time series, and the attention mechanism assigned adaptive weights to meteorological factors. The uncorrected CMA-WSP 2.0 forecast products served as a control. The effectiveness of the model correction was evaluated using metrics including the coefficient of determination ( R²), root mean square error (RMSE), and mean absolute error (MAE). The results indicate that, using 2025 data as the test set, R² of the corrected radiation forecast significantly improved, while RMSE and MAE decreased by 10.57% and 7.57%, respectively. Evaluation of March data under different sky conditions showed that the average relative errors under overcast and cloudy conditions were reduced from 28.87% and 25.47% to 3.06% and 0.33%, respectively, with skill scores (indicating the reduction in errors of the corrected forecasts relative to the original forecast values) of 89.4% and 98.7%, indicating a significant enhancement in forecasting accuracy. Under weather conditions with cloud cover and rapid radiation fluctuations, the model’s responsiveness to abrupt radiation changes was also enhanced. This study provides a theoretical basis and practical guidance for power generation forecasting at photovoltaic power stations and optimization of power grid scheduling.

关键词

CMA-WSP 2.0 / 长短期记忆(LSTM) / 注意力机制 / 光伏辐照度 / 短期预报

Key words

CMA-WSP 2.0 / long short-term memory (LSTM) / attention mechanism / photovoltaic irradiance / short-term forecasting

引用本文

引用格式 ▾
曾荣阳,雍佳,武万里,刘建宏,龚晓丽. 基于 LSTM 模型的光伏辐照度短期预报订正技术[J]. 西北工程技术学报(中英文), 2026, 25(2): 110-119 DOI:10.26974/j.cnki.XBGC.2026.02.002

登录浏览全文

4963

注册一个新账户 忘记密码

参考文献

[1]

王东风, 刘婧, 黄宇, 等 . 结合太阳辐射量计算与 CNN-LSTM 组合的光伏功率预测方法研究[J]. 太阳能学报, 2024, 45(2): 443-450.

[2]

申彦波. 新型能源体系建设中的气象问题与技术进展[J]. 南方能源建设, 2025, 12(1): 31-42.

[3]

缑晓辉, 严晓瑜, 叶冬, 等. CMA-WSP模式在宁夏太阳能预报中的适用性研究[J/OL]. 宁夏大学学报(自然科学版), 1-11(2024-09-29)[2025-10-28]. https://doi.org/10.20176/j.cnki.nxdz.000056.

[4]

万超,刘涛,曾莉萍,等. 基于CMA-WSP 1.0的贵州短波辐射检验分析[J]. 科技创新与应用,2023,13(30): 104-107.

[5]

段济开, 陈香月, 王文鹏, 等 . 基于 WRF-Solar 和 VMD-BiGRU的超短期太阳辐射订正预报研究[J]. 太阳能学报, 2025, 46(1): 710-716.

[6]

Pierro M, Bucci F, Cornaro C, et al. Model output statistics cascade to improve day ahead solar irradiance forecast[J]. Solar Energy, 2015, 117: 99-113.

[7]

顾婷婷, 潘娅英, 张加易. 浙江省中尺度数值预报系统的地表太阳辐射预报订正方法[J]. 干旱气象, 2022, 40(2): 327-332.

[8]

Lauret P, Alonso-Suárez R, Le Gal La Salle J, et al. Solar forecasts based on the clear sky index or the clearness index: Which is better?[J]. Solar, 2022, 2(4): 432-444.

[9]

Olatomiwa L, Mekhilef S, Shamshirband S, et al. A support vector machine-firefly algorithm-based model for global solar radiation prediction[J]. Solar Energy, 2015, 115: 632-644.

[10]

Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.

[11]

Chen T Q, Guestrin C. XGBoost: A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco: ACM, 2016: 785-794.

[12]

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.

[13]

Zhu T T, Li Y Z, Li Z Y, et al. Inter-hour forecast of solar radiation based on long short-term memory with attention mechanism and genetic algorithm[J]. Energies, 2022, 15(3): 1062.

[14]

Hou X X, Ju C, Wang B. Prediction of solar irradiance using convolutional neural network and attention mechanism-based long short-term memory network based on similar day analysis and an attention mechanism[J]. Heliyon, 2023, 9(11): e21484.

[15]

Son N, Jung M. Analysis of meteorological factor multivariate models for medium- and long-term photovoltaic solar power forecasting using long short-term memory[J]. Applied Sciences, 2021, 11(1): 316.

[16]

高寒旭, 袁祖晴, 张淑婷, 等. 基于LSTM 模型的短期光伏功率预测[J]. 太阳能学报, 2024, 45(6): 376-381.

[17]

Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.

[18]

Bahdanau D, Cho K , Bengio. Neural machine translation by jointly learning to align and translate[PP/OL]. arXiv(2014-09-21)[2025-10-28]. https://doi.org/10.48550/arXiv.1409.0473.

[19]

Kingma D P, Ba J. Adam: A method for stochastic optimization[PP/OL]. arXiv(2014-12-22)[2025-10-28]. https://doi.org/10.48550/arXiv.1412.6980.

[20]

王明, 孟丹, 许沛华, 等. 中国气象局风能太阳能预报系统(CMA-WSP)在风资源短期预报中的检验评估[J]. 南方能源建设, 2024, 11(1): 73-84.

基金资助

中国气象局旱区特色农业气象灾害监测预警与风险管理重点实验室项目(CAMT-202503,CAMP-202508)

宁夏自然科学基金项目(2022AAC03676)

宁夏气象局重点创新团队项目

AI Summary AI Mindmap
PDF (1548KB)

0

访问

0

被引

详细

导航
相关文章

AI思维导图

/

〈 〉