EN基于自回归与高频动态耦合模型的日典型特征光伏功率4~168 h短中期多步预测
基于自回归与高频动态耦合模型的日典型特征光伏功率4~168 h短中期多步预测
- 太阳能学报 2025年46卷第6期 页码：298-305
- 作者机构：
- 作者简介：
- 基金信息：
DOI：doi:10.19912/j.0254-0096.tynxb.2024-0277
中图分类号： TM615-
网络出版：2025-07-07，
纸质出版：2025
- 稿件说明：
移动端阅览
黄晶, 袁程旭, 郭苏. 基于自回归与高频动态耦合模型的日典型特征光伏功率4~168 h短中期多步预测[J]. 太阳能学报, 2025,46(6):298-305.
黄晶, 袁程旭, 郭苏. 基于自回归与高频动态耦合模型的日典型特征光伏功率4~168 h短中期多步预测[J]. 2025, 46(6): 298-305.
黄晶, 袁程旭, 郭苏. 基于自回归与高频动态耦合模型的日典型特征光伏功率4~168 h短中期多步预测[J]. 太阳能学报, 2025,46(6):298-305. DOI： doi:10.19912/j.0254-0096.tynxb.2024-0277.
黄晶, 袁程旭, 郭苏. 基于自回归与高频动态耦合模型的日典型特征光伏功率4~168 h短中期多步预测[J]. 2025, 46(6): 298-305. DOI： doi:10.19912/j.0254-0096.tynxb.2024-0277.
摘要
提出一种基于自回归与高频动态耦合模型的日典型特征多步预测方法。对光伏出力数据与气象信息进行相关性及数理统计分析
建立日典型特征矩阵
通过日典型特征矩阵对自回归与高频动态耦合模型进行修正
实现短期（4 h-16步、24 h-96步）
中期（168 h-168步）多步预测
结果表明该模型具有较高的预测精度
相比于长短期记忆网络（LSTM）误差降低26.1~57.8个百分点
其中提前7 d预测的全年NRMSE为21.8%。
Abstract
A daily typical characteristics multi-step prediction method based on a coupled autoregressive and dynamical system model is proposed. The correlation between PV output data and weather information is analyzed and examined
and the matrix of daily typical characteristics is established. Then
through the matrix information of daily typical characteristics
the forecasting value of coupled autoregressive and dynamical system model is fixed as exponential smoothing to cuhieve short-term （16 steps in 4 hours
96 steps in 24 hours）
medium-term （168 hours 168 steps） multi-step prediction. Comparing with the popular LSTM model
the results show that the model has a high prediction accuracy and the error is reduced by 26.1-57.8 percentage points. Of this
the full year NRMSE value predicted 7 days in advance is 21.8%.
关键词
Keywords
references
YANG D Z, WANG W T, GUEYMARD C A, et al.A review of solar forecasting, its dependence on atmospheric sciences and implications for grid integration: Towards carbon neutrality[J]. Renewable and sustainable energy reviews, 2022, 161: 112348.
NIU T, LI J K, WEI W, et al.A hybrid deep learning framework integrating feature selection and transfer learning for multi-step global horizontal irradiation forecasting[J]. Applied energy, 2022, 326: 119964.
VERBOIS H, SAINT-DRENAN Y M, THIERY A, et al. Statistical learning for NWP post-processing: a benchmark for solar irradiance forecasting[J]. Solar energy, 2022, 238: 132-149.
BELMAHDI B, LOUZAZNI M, EL BOUARDI A.One month-ahead forecasting of mean daily global solar radiation using time series models[J]. Optik, 2020, 219: 165207.
LOUZAZNI M, MOSALAM H, KHOUYA A, et al.A non-linear auto-regressive exogenous method to forecast the photovoltaic power output[J]. Sustainable energy technologies and assessments, 2020, 38: 100670.
王冉冉, 高慧敏, 张昕宇. 基于GA-GRU神经网络的光伏MPPT算法[J]. 太阳能学报, 2023, 44(9): 212-219.
GUPTA P, SINGH R.Univariate model for hour ahead multi-step solar irradiance forecasting[C]//2021 IEEE 48th Photovoltaic Specialists Conference (PVSC), Fort Lauderdale, FL, USA, 2021: 0494-0501.
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.
SHARIFI S S, REZAVERDINEJAD V, NOURANI V, et al.Multi-time-step ahead daily global solar radiation forecasting: performance evaluation of wavelet-based artificial neural network model[J]. Meteorology and atmospheric physics, 2022, 134(3): 50.
REN B W, HUANG C Q, CHEN L J, et al.CLSTM-AR-based multi-dimensional feature fusion for multi-energy load forecasting[J]. Electronics, 2022, 11(21): 3481.
李薇, 王鑫鹏, 许野, 等. 基于传递闭包的光伏短期功率组合预测方法研究[J]. 太阳能学报, 2023, 44(6): 265-274.
李丹, 张远航, 杨保华, 等. 基于约束并行LSTM分位数回归的短期电力负荷概率预测方法[J]. 电网技术, 2021, 45(4): 1356-1364.
史凯钰, 张东霞, 韩肖清, 等. 基于LSTM与迁移学习的光伏发电功率预测数字孪生模型[J]. 电网技术, 2022, 46(4): 1363-1372.
LIANG R B, QIN B B, XIA Q.Bayesian inference for mixed Gaussian GARCH-type model by Hamiltonian Monte Carlo algorithm[J]. Computational economics, 2022: 1-28.
KAKADE K, JAIN I, MISHRA A K.Value-at-risk forecasting: a hybrid ensemble learning GARCH-LSTM based approach[J]. Resources policy, 2022, 78: 102903.
LI Q, ZHANG X Y, MA T J, et al.A Multi-step ahead photovoltaic power forecasting model based on TimeGAN, Soft DTW-based K-medoids clustering, and a CNN-GRU hybrid neural network[J]. Energy reports, 2022, 8: 10346-10362.
DHAKE H, KASHYAP Y, KOSMOPOULOS P.Algorithms for hyperparameter tuning of LSTMs for time series forecasting[J]. Remote sensing, 2023, 15(8): 2076.
王福忠, 王帅峰, 张丽. 基于VMD-LSTM与误差补偿的光伏发电超短期功率预测[J]. 太阳能学报, 2022, 43(8): 96-103.
HUANG J, KOROLKIEWICZ M, AGRAWAL M, et al.Forecasting solar radiation on an hourly time scale using a coupled autoregressive and dynamical system (CARDS) model[J]. Solar energy, 2013, 87: 136-149.
HU Q H, ZHANG R J, ZHOU Y C.Transfer learning for short-term wind speed prediction with deep neural networks[J]. Renewable energy, 2016, 85: 83-95.
WHEATCROFT E.Interpreting the skill score form of forecast performance metrics[J]. International journal of forecasting, 2019, 35(2): 573-579.
GUPTA P, SINGH R.Forecasting hourly day-ahead solar photovoltaic power generation by assembling a new adaptive multivariate data analysis with a long short-term memory network[J]. Sustainable energy, grids and networks, 2023, 35: 101133.
ELIZABETH MICHAEL N, MISHRA M, HASAN S, et al.Short-term solar power predicting model based on multi-step CNN stacked LSTM technique[J]. Energies, 2022, 15(6): 2150.
BHATT A, ONGSAKUL W, NIMAL MADHU M, et al.Sliding window approach with first-order differencing for very short-term solar irradiance forecasting using deep learning models[J]. Sustainable energy technologies and assessments, 2022, 50: 101864.
GAO Y, MIYATA S, AKASHI Y.Multi-step solar irradiation prediction based on weather forecast and generative deep learning model[J]. Renewable energy, 2022, 188: 637-650.
LIMOUNI T, YAAGOUBI R, BOUZIANE K, et al.Accurate one step and multistep forecasting of very short-term PV power using LSTM-TCN model[J]. Renewable energy, 2023, 205: 1010-1024.
GUARISO G, NUNNARI G, SANGIORGIO M.Multi-step solar irradiance forecasting and domain adaptation of deep neural networks[J]. Energies, 2020, 13(15): 3987.
NOURANI V, SHARGHI E, BEHFAR N, et al.Multi-step-ahead solar irradiance modeling employing multi-frequency deep learning models and climatic data[J]. Applied energy, 2022, 315: 119069.
0
浏览量
5
下载量
0
CSCD
- Meta-XML下载
- BibTeX下载
- Endnote下载
- RefMan 下载
- NoteExpress下载
- NoteFirst下载
关联资源
相关文章
相关作者
相关机构
AI问答- 技术支持由北京北大方正电子有限公司提供 京ICP备09064830号-19
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies