Day-ahead Irradiance Forecasting Framework Based on CEEMDAN-KDBSCAN-Informer

XIE Qiyue; LIU Yao; SHEN Zhongli; ZHOU Yucai; FU Qiang; WANG Xiaoli

doi:10.19805/j.cnki.jcspe.2025.240454

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Day-ahead Irradiance Forecasting Framework Based on CEEMDAN-KDBSCAN-Informer

更新时间：2026-10-10

- Day-ahead Irradiance Forecasting Framework Based on CEEMDAN-KDBSCAN-Informer
- Chinese Journal of Power Engineering Vol. 45, Issue 9, Pages: 1433-1441(2025)
- 作者机构：
  
  1. 长沙理工大学电气与信息工程学院,湖南,长沙,410114
  2. 中南大学自动化学院,湖南,长沙,410083
- 作者简介：
- 基金信息：
- DOI：10.19805/j.cnki.jcspe.2025.240454
  CLC：
- Online First：16 September 2025，
  
  Published：2025
- 稿件说明：
移动端阅览
谢七月,刘瑶,申忠利,周育才,付强,王晓丽. 基于CEEMDAN-KDBSCAN-Informer的日前辐照度预测方法动力工程学报, 2025, 45(9): 1433-1441 https://doi.

org/10.19805/j.cnki.jcspe.2025.240454
谢七月,刘瑶,申忠利,周育才,付强,王晓丽. 基于CEEMDAN-KDBSCAN-Informer的日前辐照度预测方法动力工程学报, 2025, 45(9): 1433-1441 https://doi. DOI： 10.19805/j.cnki.jcspe.2025.240454.

org/10.19805/j.cnki.jcspe.2025.240454 DOI：

摘要

针对目前辐照度预测大多为小时内预测

且仅基于天气分类

没有考虑到一天之间天气的变化以及云类型对辐照度的影响等问题

提出一种基于完全自适应噪声集合经验模态分解(CEEMDAN)-K-means基于密度的聚类算法(KDBSCAN)-Informer的日前辐照度预测方法。首先对原始气象变量进行主成分分析来降低输入数据的维度

再通过CEEMDAN对主成分数据进行分解与重构

然后使用KDBSCAN方法对数据实现天气类型聚类和同一天气条件下云层的聚类

最后根据两层聚类结果

使用Informer模型进行日前直接法向辐照度(DNI)预测。结果表明:通过对云层类型进行分类

以在预测过程中考虑一天的天气变化

晴天日前辐照度预测的决定系数

达到0.98;相比于常用的辐照度预测方法

整体数据的平均绝对误差降低了32.42 W/m

均方根误差降低了31.71 W/m

达到0.97。

Abstract

Existing irradiance forecasts are mostly intra-hour models built on coarse weather-type labels

ignoring the intraday weather evolution and the impact of cloud type on irradiance. To address these limitations

a day-ahead direct normal irradiance (DNI) forecasting framework that couples complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN)

K-means density-based spatial clustering of applications with noise (KDBSCAN) and the Informer model was proposed. Firstly

principal component analysis was employed to compress the dimensions of raw meteorological variables. The derived principal components were then decomposed and re

constructed via CEEMDAN. KDBSCAN was subsequently employed to perform a two-layer clustering

weather-type clustering and intra-weather-type cloud-pattern clustering. Finally

based on the two-layer labels

the day-ahead DNI was forecasted via the Informer network. Results show that by classifying cloud types to consider weather changes throughout the day in the prediction process

the determination coefficient (

) for predicting irradiance before sunny days reaches 0.98. Compared with prevailing benchmarks

the proposed framework reduces the overall mean absolute error by 32.42 W/m

and root-mean-square error by 31.71 W/m

achieving overall

of 0.97.

关键词

Keywords

references

HOU Xinxing, JU Chao, WANG Bo. 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.

DOU Weijing, WANG Kai, SHAN Shou, et al. Day-ahead numerical weather prediction solar irradiance correction using a clustering method based on weather conditions[J]. Applied Energy, 2024, 365: 123239.

ZHU Tingting, ZHOU Hai, WEI Hankun, et al. Inter-hour direct normal irradiance forecast with multiple data types and time-series[J]. Journal of Modern Power Systems and Clean Energy, 2019, 7(5): 1319-1327.

KAROUT Y, THIL S, EYNARD J, et al. Hybrid intrahour DNI forecast model based on DNI measurements and sky-imaging data[J]. Solar Energy, 2023, 249: 541-558.

孙立, 王显连, 苏志刚, 等. 计及源荷不确定性的光/氢/储建筑能源系统高效热电协同智能调度[J]. 工程热物理学报, 2024, 45(7): 1932-1940. SUN Li, WANG Xianlian, SU Zhigang, et al. Intelligent scheduling for efficient thermal-electric coordination of a photovoltaic/hydrogen/storage building energy system considering source-load uncertainties[J]. Journal of Engineering Thermophysics, 2024, 45(7): 1932-1940.

NOURI B, BLUM N, WILBERT S, et al. A hybrid solar irradiance nowcasting approach: combining all sky imager systems and persistence irradiance models for increased accuracy[J]. Solar RRL, 2022, 6(5): 2100442.

朱琼锋, 李家腾, 乔骥, 等. 人工智能技术在新能源功率预测的应用及展望[J]. 中国电机工程学报, 2023, 43(8): 3027-3047. ZHU Qiongfeng, LI Jiateng, QIAO Ji, et al. Application and prospect of artificial intelligence technology in renewable energy forecasting[J]. Proceedings of the CSEE, 2023, 43(8): 3027-3047.

ALLAL Z, NOURA H N, CHAHINE K. Machine learning algorithms for solar irradiance prediction: a recent comparative study[J]. e-Prime-Advances in Electrical Engineering, Electronics and Energy, 2024, 7: 100453.

LI Xianglong, MA Longfei, CHEN Ping, et al. Probabilistic solar irradiance forecasting based on XGBoost[J]. Energy Reports, 2022, 8: 1087-1095.

QING Xiangyun, NIU Yugang. Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM[J]. Energy, 2018, 148: 461-468.

董坤, 冉鹏, 刘旭, 等. 基于权参数优化的并行深度学习光伏功率预测[J]. 动力工程学报, 2024, 44(1): 91-98. DONG Kun, RAN Peng, LIU Xu, et al. Photovoltaic power prediction by weight parameter optimization-based parallel deep learning[J]. Journal of Chinese Society of Power Engineering, 2024, 44(1): 91-98.

WANG Hui, WANG Peng, ZHANG Yiyi, et al. Short-term prediction of solar irradiance based on EEMD-Lasso-MLP[C]//Proceedings of 2022 International Conference on Machine Learning and Knowledge Engineering (MLKE). Guilin: IEEE, 2022: 164-168.

臧海祥, 张越, 程礼临, 等. 基于ICEEMDAN-LSTM和残差注意力的短期太阳辐照度预测[J]. 太阳能学报, 2023, 44(12): 175-181. ZANG Haixiang, ZHANG Yue, CHENG Lilin, et al. Short term solar radiation forecasting based on ICEEMDAN-LSTM and residual attention[J]. Acta Energiae Solaris Sinica, 2023, 44(12): 175-181.

闫文杰, 徐志杰, 薛桂香, 等. 基于EMD-TCN的多能供热系统太阳辐照度预测模型[J]. 太阳能学报, 2023, 44(11): 182-188. YAN Wenjie, XU Zhijie, XUE Guixiang, et al. Solar irradiance forecasting model based on EMD-tcn for multi-energy heating systems[J]. Acta Energiae Solaris Sinica, 2023, 44(11): 182-188.

XIE Qiyue, MA Lin, LIU Yao, et al. An improved SSA-BiLSTM-based short-term irradiance prediction model via sky images feature extraction[J]. Renewable Energy, 2023, 219: 119507.

杨家豪, 张莲, 梁法政, 等. 基于天气二次分类的地表太阳辐射预测方法[J]. 分布式能源, 2024, 9(1): 54-63. YANG Jiahao, ZHANG Lian, LIANG Fazheng, et al. Surface solar radiation forecast using quadratic weather classification method[J]. Distributed Energy, 2024, 9(1): 54-63.

张玉, 王瑜琳. 基于天气聚类的光伏电站辐照度预测[J]. 科学技术与工程, 2021, 21(3): 1030-1036. ZHANG Yu, WANG Yulin. The prediction of irradiation degree of photovoltaic power station based on weather clustering[J]. Science Technology and Engineering, 2021, 21(3): 1030-1036.

CHENG H Y, YU C C. Multi-model solar irradiance prediction based on automatic cloud classification[J]. Energy, 2015, 91: 579-587.

肖俊青, 金江涛, 岳敏楠, 等. 改进CEEMDAN算法与分形融合的深度学习轴承故障分析[J]. 动力工程学报, 2022, 42(6): 522-529. XIAO Junqing, JIN Jiangtao, YUE Minnan, et al. Bearing fault analysis of deep learning based on improved CEEMDAN algorithm and fractal fusion[J]. Journal of Chinese Society of Power Engineering, 2022, 42(6): 522-529.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Boiler Reheater Wall Temperature Prediction Model Based on CNN-BiLSTM-SSA

Mooring-line Tension Prediction Based on Re-decomposition and Deep Learning

Study on Fault Early Warning of Induced Draft Fan Based on TCN-SDAE Model and Adaptive Dynamic Threshold

Prediction of SO2 Emission Concentration for CFB Units Under Quick Load Change Conditions Based on MLO-HIKAN

Spatio-temporal Graph Attention Network-based Fault Warning of Coal Mill

Related Author

XU Shiming

HE Zhiqian

PENG Xianyong

SHANG Zhongbao

FAN Jingwei

WANG Junlue

QU Shuyang

LIU Yang

Related Institution

School of Low Carbon Energy and Power Engineering, China University of Mining and Technology

Huaneng Yingkou Thermal Power Co., Ltd.

Jiangsu Smart Energy Technology and Equipment Engineering Research Center

Non-carbon Energy Conversion and Utilization Institute

School of Energy and Power Engineering, University of Shanghai for Science and Technology

AI问答

⁰