
1. 贵州电网有限责任公司毕节供电局,贵州,毕节,551700
2. 重庆邮电大学 通信与信息工程学院,重庆,400065
纸质出版:2025
移动端阅览
刘庆,王有军,张垚,姜继彬,李康,李璘,王平.基于注意力机制的融合式NCP-DCNN短期光伏功率预测方法[J].智慧电力,2025,53(8):62-69.
LIU Qing, WANG Youjun, ZHANG Yao, et al. Attention Mechanism-based Integrated NCP-DCNN Method for Short-Term Photovoltaic Power Forecasting[J]. 2025, 53(8): 62-69.
刘庆,王有军,张垚,姜继彬,李康,李璘,王平.基于注意力机制的融合式NCP-DCNN短期光伏功率预测方法[J].智慧电力,2025,53(8):62-69. DOI: 10.20204/j.sp.2025.08008.
LIU Qing, WANG Youjun, ZHANG Yao, et al. Attention Mechanism-based Integrated NCP-DCNN Method for Short-Term Photovoltaic Power Forecasting[J]. 2025, 53(8): 62-69. DOI: 10.20204/j.sp.2025.08008.
针对传统短期光伏功率预测模型普遍存在预测精度不足、参数量过大等问题,提出一种融合注意力机制(AM)、神经回路策略(NCP)、密集卷积神经网络(DCNN)的短期光伏功率预测方法。首先,引入皮尔逊相关分析识别影响光伏发电的关键因素,并将其作为网络输入提高特征学习效率;其次,通过多层卷积操作提取输入数据的局部空间特征,并利用自注意力机制自动聚焦于重要的时间段和区域,增强特征表征能力;最后,结合神经回路策略模块实现特征图在时间维度的特征提取。仿真结果表明,所提方法在不依赖网络参数的情况下具有较高的短期预测精度,可显著降低网络复杂度并提升计算效率。
To address the common issues of insufficient prediction accuracy and excessive parameters in traditional short-term photovoltaic (PV) power forecasting models, this study proposes a novel method integrating attention mechanism (AM), neural circuit policy (NCP), and densely connected convolutional neural network (DCNN). First, Pearson correlation analysis is introduced to identify key factors influencing PV power generation, which are then used as network inputs to enhance feature learning efficiency. Second, multi-layer convolutional operations extract local spatial features from input data, while a self-attention mechanism automatically focuses on critical time segments and regions to strengthen feature representation. Finally, a NCP module enables feature map extraction along the temporal dimension. Simulation results demonstrate that the proposed method achieves high short-term forecasting accuracy without relying on network parameter expansion, significantly reducing network complexity and improving computational efficiency.
张翼,周超.城市轨道交通车站出入口分布式光伏发电应用研究[J].城市轨道交通研究, 2024,27(6):291-295.
王晖,赵咨钧,管保晋,等.光伏发电机组异动信息主动增量式更新算法[J].电子设计工程,2025,33(1):132-136.
杨涛,严大洲,温国胜,等.新能源产业链构建:光伏发电-电化学储能-新能源汽车[J].中国材料进展,2024,43(2):164-174.
陆毅,薛枫,唐小波,等.基于余弦相似度和TSO-BP的短期光伏预测方法[J].浙江电力,2024,43(6):22-30.
吴世健,廖强明,邱建东.基于大数据分析的分布式光伏预测与控制策略[J].电器工业,2024(6):45-49.
PEDRO H,COIMBRA C.Assessment of forecasting techniques for solar power production with no exogenous inputs[J].Solar Energy,2012,86(7):2017-2028.
DIAGNE M,DAVID M,LAURET P,et al.Review of solar irradiance forecasting methods and a proposition for small-scale insular grids[J].Renewable and Sustainable Energy Reviews,2013,27(11):65-76.
陈习勋,吴凯彤,何杰,等.基于集成机器学习模型的短期光伏出力区预测[J].智慧电力,2024,52(2):87-93.
刘源延.基于机器学习的短期光伏发电功率预测[D].北京:华北电力大学,2023.
HUTHIAH H,SA U,EFENDI A.Support vector regression (SVR) model for seasonal time series data[C]//2021 Second Asia Pacific International Conference on Industrial Engineering and Operations Management (APCIEOM).Wenzhou,China,2021:3191-3200.
邹港,赵斌,罗强,等.基于PCA ‑ VMD ‑ MVO ‑ SVM的短期光伏输出功率预测方法[J].电力科学与技术学报,2024,39(5):163-171.
MOORE P J,LYONS T J,GALLACHER J.Random forest prediction of Alzheimer’s disease using pairwise selection from time series data[J].PLOS ONE,2019,14(2):1-14.
张程珂,刘会灯,朱渝宁,等.基于多特征分析提取的随机森林超短期光伏功率预测[J].电力需求侧管理,2023,25(6):50-56.
贾凌云,云斯宁,赵泽妮,等.神经网络短期光伏发电预测的应用研究进展[J].太阳能学报,2022,43(12):88-97.
蔡昌春,范靖浩,李源佳,等.基于TPA‑MBLSTM模型的超短期风电功率预测[J].电力科学与技术学报,2024,39(1):47-56.
李明,袁逸萍,贾依达尔,等.考虑PCA-LSTM的风电机组输出功率预测研究[J].机械设计与制造,2022,379(9):145-148.
孟鑫禹,王睿涵,张喜平,等.基于经验模态分解与多分支神经网络的超短期风功率预测[J].计算机应用,2021,41(1):237-242.
周磊,竺筱晶.基于MA-CNN-LSTM和自注意力机制的单变量短期电力负荷预测[J].科学技术与工程,2024,24(22):9408-9416.
马磊,黄伟,李克成,等.基于Attention-LSTM的光伏超短期功率预测模型[J].电测与仪表,2021,58(2):146-152.
ZHOU H X,ZHANG Y J,YANG L F,et al.Short-term photovoltaic power forecasting based on long short-term memory neural network and attention mechanism[J].IEEE Access,2019,7(99):78063-78074.
DING Y K,ZHU Y L,WU Y R,et al.Spatio-temporal attention LSTM model for flood forecasting[C]//2019 International Conference on Internet of Things (iThings).Atlanta,USA,2019:458-465.
LECHNER M,HASANI R,AMINI A,et al.Neural circuit policies enabling auditable autonomy[J].Nature Machine Intelligence,2020,1:642-652.
MA D,GUO Y,MA S.Short-term subway passenger flow prediction based on GCN-BiLSTM[C]//2020 IOP Conference Series:Earth and Environmental Science.Xi’an,China,2020.
侯慧,吴文杰,魏瑞增,等.基于注意力机制的CNN-LSTM-XGBoost台风暴雨电力气象混合预测模型[J].智慧电力,2024,52(10):96-102.
WANG Q,WU B,ZHU P,et al.ECA-net:efficient channel attention for deep convolutional neural networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Seattle,USA,2020:11531-11539.
ZHANG H,ZU K K,LU J,et al.EPSANet:an efficient pyramid squeeze attention block on convolutional neural network[C]//16th Asian Conference on Computer Vision Macao,China,2022:1161-1177.
HUANG G,LIU Z,MAATEN L V D,et al.Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Honolulu,USA,2017:4700-4708.
HASANI R,LECHNER M,AMINI A,et al.Liquid time-constant networks[C]//2020 AAAI Conf.Artif.Intell.New York,USA,2020:7657-7666.
0
浏览量
31
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621