基于BSimilar优化PTransformer的光伏功率短期预测

张文广; 蔡浩; 刘科; 孙盼荣

doi:10.19805/j.cnki.jcspe.2026.240722

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于BSimilar优化PTransformer的光伏功率短期预测

新能源与储能 | 更新时间：2026-10-10

- 基于BSimilar优化PTransformer的光伏功率短期预测
- Short-term Prediction of PV Power Based on BSimilar Optimized PTransformer
- 动力工程学报 2026年46卷第1期页码：77-84
- 作者机构：
  
  1. 华北电力大学新能源电力系统全国重点实验室,北京,102206
  2. 华北电力大学控制与计算机工程学院,北京,102206
- 作者简介：
  
  张文广(1975—),男,山东烟台人,教授,博士生导师,博士,研究方向为新能源发电功率预测及机组智能故障诊断与运维等
  蔡浩(通信作者),男,硕士研究生,E-mail:caihao0212@163.com
- 基金信息：
  
  华能集团总部科技项目(HNKJ20-H88-05)
- DOI：10.19805/j.cnki.jcspe.2026.240722
  中图分类号：
- 网络首发：2026-01-15，
  
  纸质出版：2026-01-15
- 稿件说明：
移动端阅览
张文广,蔡浩,刘科,孙盼荣. 基于BSimilar优化PTransformer的光伏功率短期预测动力工程学报, 2026, 46(1): 77-84 https://doi.

org/10.19805/j.cnki.jcspe.2026.240722
张文广,蔡浩,刘科,孙盼荣. 基于BSimilar优化PTransformer的光伏功率短期预测动力工程学报, 2026, 46(1): 77-84 https://doi. DOI： 10.19805/j.cnki.jcspe.2026.240722.

org/10.19805/j.cnki.jcspe.2026.240722 DOI：

摘要

为提高光伏功率短期预测的精度

提出了考虑光伏设备性能退化因素的相似日算法优化的分时段多通道独立光伏功率短期预测方法。首先

在PTransformer模型中用分时段与通道独立的方法来处理光伏输入数据

以降低空间复杂度及提高长时间数据序列的关注度。其次

运用Transformer的编码器模型

通过自身注意力机制捕捉光伏序列特征之间的依赖关系

进行光伏功率的短期预测。最后

运用夹角余弦距离计算相似度并考虑光伏设备性能退化因素确定相似日

利用其功率数据优化PTransformer模型

以改善功率数据的滞后性。结果表明:相比典型的光伏功率短期预测方法

所提方法训练速度更快

预测精准度更高

并且对复杂天气状况下的光伏功率也有较好的预测结果。

Abstract

To enhance the accuracy of short-term photovoltaic power forecasting

a segmented multi-channel independent short-term photovoltaic power prediction method optimized by a similar day algorithm with consideration of photovoltaic equipment performance degradation factors was proposed. Firstly

the photovoltaic input data were processed in PTransformer model with time period and channel independent methods to reduce the spatial complexity and improve the attention of long-term data series. Secondly

the encoder model of Transformer was applied to capture the dependencies between photovoltaic sequence features through its own attention mechanism for short-term photovoltaic power prediction. Finally

the PTransformer model was optimized using its power data to improve the lag of the power data by applying the angle cosine distance to calculate the similarity and considering the performance degradation factors of the photovoltaic equipment to determine the similar day. Results show that compared with typical short-term photovoltaic power prediction method

the proposed method has faster training speed

higher prediction accuracy and better prediction results for photovoltaic power under complex weather conditions.

关键词

Keywords

references

陆王琳, 李政, 姜山. 分体式太阳能PVT热泵系统试验运行性能分析[J]. 动力工程学报, 2024, 44(10): 1574-1581.LU Wanglin, LI Zheng, JIANG Shan. Experimental operation performance analysis on a solar PVT heat pump system with combined modules[J]. Journal of Chinese Society of Power Engineering, 2024, 44(10): 1574-1581.

国家统计局. 中华人民共和国2023年国民经济和社会发展统计公报[R]. 北京: 中国统计局, 2024.

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.

董存, 王铮, 白捷予, 等. 光伏发电功率超短期预测方法综述[J]. 高电压技术, 2023, 49(7): 2938-2951. DONG Cun, WANG Zheng, BAI Jieyu, et al. Review of ultra-short-term forecasting methods for photovoltaic power generation[J]. High Voltage Engineering, 2023, 49(7): 2938-2951.

张姗, 冬雷, 纪德洋, 等. 基于NWP相似性分析的超短期光伏发电功率预测[J]. 太阳能学报, 2022, 43(4): 142-147.ZHANG Shan, DONG Lei, JI Deyang, et al. Power forecasting of ultra-short-term photovoltaic station based on NWP similarity analysis[J]. Acta Energiae Solaris Sinica, 2022, 43(4): 142-147.

陈志宝, 丁杰, 周海, 等. 地基云图结合径向基函数人工神经网络的光伏功率超短期预测模型[J]. 中国电机工程学报, 2015, 35(3): 561-567.CHEN Zhibao, DING Jie, ZHOU Hai, et al. A model of very short-term photovoltaic power forecasting based on ground-based cloud images and RBF neural network[J]. Proceedings of the CSEE, 2015, 35(3): 561-567.

李超然, 潘鹏程, 杨伟荣, 等. 基于改进相似日优化HBA-BiLSTM-KELM的光伏发电功率预测[J]. 太阳能学报, 2024, 45(5): 508-516.LI Chaoran, PAN Pengcheng, YANG Weirong, et al. Research on PV system power prediction based on improved similar day and HBA-BiLSTM-KELM neural network[J]. Acta Energiae Solaris Sinica, 2024, 45(5): 508-516.

ZHOU Haoyi, ZHANG Shanghang, PENG Jieqi, et al. Informer: beyond efficient transformer for long sequence time-series forecasting[C]//Proceedings of the 35th AAAI Conference on Artificial Intelligence. Honolulu, USA: AAAI, 2021: 11106-11115.

WU Haixu, XU Jiehui, WANG Jianmin, et al. Autoformer: decomposition transformers with auto-correlation for long-term series forecasting[C]// Proceedings of the 35th International Conference on Neural Information Processing Systems. Red Hook, USA: Curran Associates Inc., 2021: 22419-22430.

黄莉, 甘恒玉, 刘兴举, 等. 基于Transformer编码器的超短期光伏发电功率预测[J]. 智慧电力, 2024, 52(5): 16-22, 59.HUANG Li, GAN Hengyu, LIU Xingju, et al. Ultra-short-term photovoltaic power generation prediction based on transformer encoder[J]. Smart Power, 2024, 52(5): 16-22, 59.

LI Shiyang, JIN Xiaoyong, XUAN Yao, et al. Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting[C]// Proceedings of the 33rd International Conference on Neural Information Processing Systems. Red Hook, USA: Curran Associates Inc., 2019: 471.

YANG Jing, HE Hui, ZHAO Xuemeng, et al. Day-ahead PV power forecasting model based on fine-grained temporal attention and cloud-coverage spatial attention[J]. IEEE Transactions on Sustainable Energy, 2024, 15(2): 1062-1073.

NIE Yuqi, NGUYEN N H, SINTHONG P, et al. A time series is worth 64 words: long-term forecasting with transformers[C]// International Conference on Learning Representations. Kigali: ICLR, 2023.

杨茂, 孟玲建, 李大勇, 等. 基于类3准则的光伏功率异常数据识别[J]. 可再生能源, 2018, 36(10): 1443-1448. YANG Mao, MENG Lingjian, LI Dayong, et al. Identification of abnormal data of photovoltaic power based on class 3[J]. Renewable Energy Resources, 2018, 36(10): 1443-1448.

MISHRA M, BYOMAKESHA DASH P, NAYAK J, et al. Deep learning and wavelet transform integrated approach for short-term solar PV power prediction[J]. Measurement, 2020, 166: 108250.

王言文, 邱启荣, 王宝坤. 基于Lasso和SVR的向量夹角余弦变权重组合预测模型[J]. 统计与决策, 2020, 36(18): 22-26.WANG Yanwen, QIU Qirong, WANG Baokun. Forecasting model of vector angle cosine variable weight combination based on lasso and support vector regression[J]. Statistics & Decision, 2020, 36(18): 22-26.

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

基于BKA-Transformer-BiGRU的垃圾焚烧炉SNCR脱硝系统动态建模

基于WOA的循环流化床锅炉SO2排放浓度VMD-Transformer预测模型

煤-氨混燃锅炉建模及近邻传播-最小支持向量机多模态验证优化

基于T-S模糊模型的风力机系统事件触发控制

分形特征对漂浮式风力机仿生平台动态响应的影响