基于TimeVAE的1DCNN-S-Mamba组合模型光伏功率短期预测

许可证; 文中; 王秋杰

doi:10.19666/j.rlfd.202506097

欢迎来到能源电力期刊网！

您当前的位置：

首页 >

文章列表页 >

基于TimeVAE的1DCNN-S-Mamba组合模型光伏功率短期预测

热能科学研究 | 更新时间：2026-03-18

- 基于TimeVAE的1DCNN-S-Mamba组合模型光伏功率短期预测
- 基于TimeVAE的1DCNN-S-Mamba组合模型光伏功率短期预测
- 热力发电 2026年55卷第1期页码：122-133
- 作者机构：
  
  1. 三峡大学电气与新能源学院
  2. 新能源微电网湖北省协同创新中心(三峡大学)
- 作者简介：
- 基金信息：
  
  湖北省自然科学基金创新发展联合基金项目（2024AFD362）~~
- DOI：10.19666/j.rlfd.202506097
  中图分类号： TM615;TP18
- 纸质出版：2026
- 稿件说明：
移动端阅览
[1]许可证,文中,王秋杰.基于TimeVAE的1DCNN-S-Mamba组合模型光伏功率短期预测[J].热力发电,2026,55(01):122-133.
[1]许可证,文中,王秋杰.基于TimeVAE的1DCNN-S-Mamba组合模型光伏功率短期预测[J].热力发电,2026,55(01):122-133. DOI： 10.19666/j.rlfd.202506097.

DOI：10.19666/j.rlfd.202506097.

摘要

针对极端天气下光伏功率预测存在的气象响应失准、突变特征捕捉困难及数据稀缺等问题，提出一种基于模糊C均值（fuzzy C-means

FCM）、最大信息系数（maximum information coefficient

MIC）、时序变分自编码器（time variational auto-encoders

TimeVAE）、一维卷积神经网络（1D convolutional neural network

1DCNN）和simple-Mamba(S-Mamba)的组合功率预测模型。首先，通过气象特征结合FCM聚类将天气划分为晴天、多云、降雪和降雨4类；然后，结合MIC筛选出最佳气象特征子集，同时针对极端天气样本匮乏问题，采用Time VAE进行数据生成，利用其分解式重构机制生成仿真数据；最后，使用1DCNN-S-Mamba组合模型通过局部卷积捕获短时突变特征，结合双向状态空间建模实现长程依赖解析进行预测。实验结果表明，该模型提升了复杂天气下光伏功率预测的时效性与准确性。相较于S-Mamba，所提模型平均绝对误差和均方根误差在降雪天气下分别降低了3.65%和5.10%。

Abstract

To address the challenges of meteorological-power response inaccuracy

difficulty in capturing abrupt features

and data scarcity in photovoltaic power prediction under extreme weather conditions

a hybrid prediction framework is proposed based on fuzzy C-means(FCM)

maximum information coefficient(MIC)

time variational auto-encoders(TimeVAE)

1D convolutional neural network(1 DCNN)

and simple-Mamba(S-Mamba). Firstly

meteorological features are clustered using FCM to categorize weather into four types: sunny

cloudy

snowy

and rainy. Subsequently

MIC is employed to select the optimal subset of meteorological features. To mitigate the scarcity of extreme weather samples

Time VAE is adopted for data generation

leveraging its decomposed reconstruction mechanism to synthesize realistic time-series data. Finally

a 1 DCNN-S-Mamba combined model is utilized

where 1 DCNN captures short-term abrupt features through local convolution

while bidirectional statespace modeling in S-Mamba enables long-range dependency analysis for prediction. Experimental results demonstrate that the proposed model enhances both timeliness and accuracy in PV power prediction under complex weather conditions. Compared to S-Mamba

it reduces the mean absolute error(MAE) and root mean square error(RMSE) by 3.65% and 5.10%

respectively

in snowy weather scenarios.

关键词

Keywords

references

李国荣,张云鹏,周海,等.基于有效辐照度修正的光伏电池模型参数辨识和输出特性建模研究[J].电工技术学报, 2024, 39(23):7331-7340.LI Guorong, ZHANG Yunpeng, ZHOU Hai, et al.Parameter identification and output characteristics modelling of photovoltaic module based on effective irradiance correction[J]. Transactions of China Electrotechnical Society, 2024, 39(23):7331-7340.

陈思宇,蒋俊霞,李帅兵.极端气候条件下光伏发电功率预测方法综述[J].电气工程学报, 2025, 20(1):281-290.CHEN Siyu, JIANG Junxia, LI Shuaibing, et al. Review of photovoltaic power prediction methods under extreme climate conditions[J]. Chinese Journal of Electrical Engineering, 2025, 20(1):281-290.

叶林,裴铭,李卓,等.风电和光伏发电功率联合预测与预调度框架[J].高电压技术, 2024, 50(9):3823-3836.YE Lin, PEI Ming, LI Zhuo, et al. Framework for Joint wind and photovoltaic power forecasting and predispatch[J]. High Voltage Engineering, 2024, 50(9):3823-3836.

AKPULUMA D, YURCHENKO A V, FIROZ N, et al.Enhanced forecasting of PV power output using LSTM integrated with LASSO-RFR hybrid models in extreme weather conditions[C]//Proceedings of the 2024 IEEE 3rd International Conference on Problems of Informatics,Electronics and Radio Engineering(PIERE), 2024:780-785.

程礼临,臧海祥,卫志农,等.考虑多光谱卫星遥感的区域级超短期光伏功率预测[J].中国电机工程学报,2022, 42(20):7451-7465.CHENG Lilin, ZANG Haixiang, WEI Zhinong, et al.Ultra-short-term forecasting of regional photovoltaic power generation considering multispectral satellite remote sensing data[J]. Proceedings of the CSEE, 2022,42(20):7451-7464.

王迎春,王昱栋,刘洋,等.基于卫星遥感多光谱云图的生成式海上超短期光伏功率预测[J].控制与决策,2025, 40(4):1136-1144.WANG Yingchun, WANG Yudong, LIU Yang, et al. Ultrashort-term prediction of offshore photovoltaic power based on satellite remote sensing multi-spectral cloud image[J]. Control and Decision, 2025, 40(4):1136-1144.

陈殿昊,臧海祥,蒋雨楠,等.基于多层次云图特征与宽度学习的超短期光伏功率预测[J].电力系统自动化,2024, 48(22):131-139.CHEN Dianhao, ZANG Haixiang, JIANG Yunan, et al.Ultra-short-term photovoltaic power forecasting based on multi-level sky image features and broad learning[J].Automation of Electric Power Systems, 2024, 48(22):131-139.

刘翼飞,崔承刚.基于地基云图云量特征的光伏发电功率区间预测[J].南方电网技术, 2023, 17(2):92-100.LIU Yifei, CUI Chenggang. Interval prediction for shortterm solar power based on cloud features of ground-based cloud images[J]. Southern Power System Technology,2023, 17(2):92-100.

尹杰,刘博,孙国兵,等.基于迁移学习和降噪自编码器-长短时间记忆的锂离子电池剩余寿命预测[J].电工技术学报, 2024, 39(1):289-302.YIN Jie, LIU Bo, SUN Guobing, et al. Transfer learning denoising autoencoder-long short term memory for remaining useful life prediction of Li-ion batteries[J].Transactions of China Electrotechnical Society, 2024,39(1):289-302.

魏书荣,殷世杰,闫梦飞,等.基于改进生成对抗网络的海上风电机组故障数据增强及诊断[J].电力系统保护与控制, 2025, 53(1):114-124.WEI Shurong, YIN Shijie, YAN Mengfei, et al. Offshore wind turbine fault data enhancement and diagnosis based on an improved generative adversarial network[J]. Power System Protection and Control, 2025, 53(1):114-124.

殷豪,张铮,丁伟锋,等.基于生成对抗网络和LSTM-CSO的少样本光伏功率短期预测[J].高电压技术, 2022, 48(11):4342-4351.YIN Hao, ZHANG Zheng, DING Weifeng, et al. Shortterm prediction of small-sample photovoltaic power based on generative adversarial network and LSTM-CSO[J].High Voltage Engineering, 2022, 48(11):4342-4351.

JIANG J, KIM D, KIM B, et al. GA-VAE:enhancing local feature representation in VQ-VAE through genetic algorithm-based token optimization[J]. IEEE ACCESS,2025, 13:34286-34295.

刘君,邓毅,杨延西,等.基于条件生成对抗网络的空气预热器内红外补光监测视频图像清晰化方法[J].热力发电, 2021, 50(10):130-134.LIU Jun, DENG Yi, YANG Yanxi, et al. Method for sharpening infrared compensation image for monitoring video inside air preheater based on c GAN network[J].Thermal Power Generation, 2021, 50(10):130-134.

KIM W, JEON J, KIM S, et al. Prediction of index futures movement using time-GAN and 3D-CNN:empirical evidence from Korea and the United States[J]. Applied Soft Computing, 2025, 171:112748.

SURESH V, JANIK P, REZMER J, et al. Forecasting solar PV output using convolutional neural networks with a sliding window algorithm[J]. Energies, 2020, 13(3):723.

YU Y J, CAO J F, ZHU J Y. An LS-TM short-term solar irradiance forecasting under complicated weather conditions[J]. IEEE ACCESS, 2019, 7:145651-145666.

王家乐,张耀,林帆,等.基于自注意力特征提取的光伏功率组合概率预测[J].太阳能学报, 2024, 45(12):123-131.WANG Jiale, ZHANG Yao, LIN Fan, et al. Combining probabilistic prediction of PV power based on selfattention feature extraction mechanism[J]. Acta Energiae Solaris Sinica, 2024, 45(12):123-131.

李永飞,张耀,林帆,等.基于气候特征分析及改进XGBoost算法的中长期光伏电站发电量预测方法[J].电力系统保护与控制, 2024, 52(11):84-92.LI Yongfei, ZHANG Yao, LIN Fan, et al. Medium-and long-term power generation forecast based on climate characterisation and an improved XGBoost algorithm for photovoltaic power plants[J]. Power System Protection and Control, 2024, 52(11):84-92.

IOANNIS K, BAZIONIS, MARKOS A, et al.Investigation of LSTM-based hybrid approaches for short-term PV power forecasting[J]. IET Conference Proceedings, 2025, 2024(29):346-351.

赵如意,王晓辉,郑碧煌,等.基于特征优化和混合改进灰狼算法优化BiLSTM网络的短期光伏功率预测[J].电网技术, 2025, 49(1):209-222.ZHAO Ruyi, WANG Xiaohui, ZHENG Bihuang, et al.Short-term photovoltaic power prediction based on feature optimization and hybrid improved grey wolf algorithmoptimized BiLSTM network[J]. Power System Technology, 2025, 49(1):209-222.

林江,蔡晓龙,周剑桥.基于Transformer模型的光伏系统故障分类与预测方法[J].智慧电力, 2025, 53(4):96-102.LIN Jiang, CAI Xiaolong, ZHOU Jianqiao. Transformer model-based fault classification and prediction method for photovoltaic systems[J]. Smart Power, 2025, 53(4):96-102.

常青松,杨昭,杨熠辉,等.基于相似日聚类的超短期光伏功率组合预测模型[J].热力发电, 2023, 52(11):123-131.CHANG Qingsong, YANG Zhao, YANG Yihui, et al.Ultrashort term photovoltaic power combinatorial forecasting model based on similar day clustering[J].Thermal Power Generation, 2023, 52(11):123-131.

周育才,肖添,谢七月,等.基于聚类的HPOBILSTM光伏功率短期预测[J].太阳能学报, 2024,45(4):512-518.ZHOU Yucai, XIAO Tian, XIE Qiyue, et al. Clusteringbased HPO-BILSTM short-term prediction of PV power[J]. Acta Energiae Solaris Sinica, 2024, 45(4):512-518.

许越,李强,崔晖.基于MIC-EEMD-改进Informer的含高比例清洁能源与储能的电力市场短期电价多步预测[J].电网技术, 2024, 48(3):949-958.XU Yue, LI Qiang, CUI Hui. Short-term multi-step price prediction for the electricity market with a high proportion of clean energy and energy storage based on MIC-EEMDimproved informer[J]. Power System Technology, 2024,48(3):949-958.

WANG Z H, KONG F H, FENG S, et al. Is mamba effective for time series forecasting?[J]. Neurocomputing, 2025, 619:129178.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据