基于Q学习组合模型的光伏电站发电功率预测方法研究

唐仲杰; 寇文珍; 喻雪

doi:10.19912/j.0254-0096.tynxb.2024-1153

PDF(1055 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (9) : 618-625. DOI: 10.19912/j.0254-0096.tynxb.2024-1153

基于Q学习组合模型的光伏电站发电功率预测方法研究

唐仲杰¹, 寇文珍², 喻雪²

作者信息 +

RESEARCH ON POWER GENERATION FORECASTING METHOD FOR PHOTOVOLTAIC PLANTS BASED ON Q-LEARNING COMBINED MODEL

Tang Zhongjie¹, Kou Wenzhen², Yu Xue²

Author information +

文章历史 +

摘要

针对单一模型在预测光伏电站发电功率易受自身固有局限影响导致的预测精度不理想以及传统组合模型权重恒定非最优而未充分挖掘其预测潜力的问题,提出基于Q学习组合模型的预测方法。首先,通过Pearson相关系数选取与发电功率有强相关性的气象特征和时间序列作为输入特征;其次,分别采用极限梯度提升和长短期记忆网络两个单一模型对发电功率进行预测;最后,采用Q学习算法对历史样本在线滚动优化实时更新每次初步预测的组合权重,从而实现最优时变权重组合。结果表明,相比于其他模型,所提方法能显著提升预测精度和效果,验证了该方法的有效性。

Abstract

Aiming at the problem that the forecast of a single model in forecasting the generation power of photovoltaic power plants is not ideal due to its inherent limitations, and that the weight of the traditional combination model is constant and not optimal, and its forecast potential is not fully exploited, a prediction method based on the Q-learning combination model is proposed. Firstly, meteorological features and time series with strong correlation with power generation are selected as input features through Pearson correlation coefficient; Secondly, two single models namely XGBoost（eXtreme Gradient Boosting） and LSTM（Long Short Term Memory） are used to forecast the generation power; Finally, the Q-learning algorithm is used to perform online rolling optimization on historical samples and update the combination weights of each preliminary forecast in real-time, thereby achieving the optimal time-varying weight combination. The experimental results show that compared to other models, the method can significantly improve forecast accuracy and effectiveness, verifying the effectiveness of this method.

导出引用

唐仲杰, 寇文珍, 喻雪. 基于Q学习组合模型的光伏电站发电功率预测方法研究[J]. 太阳能学报. 2025, 46(9): 618-625 https://doi.org/10.19912/j.0254-0096.tynxb.2024-1153

Tang Zhongjie, Kou Wenzhen, Yu Xue. RESEARCH ON POWER GENERATION FORECASTING METHOD FOR PHOTOVOLTAIC PLANTS BASED ON Q-LEARNING COMBINED MODEL[J]. Acta Energiae Solaris Sinica. 2025, 46(9): 618-625 https://doi.org/10.19912/j.0254-0096.tynxb.2024-1153

中图分类号： TM615

参考文献

[1] 郭剑波, 王铁柱, 罗魁, 等. 新型电力系统面临的挑战及应对思考[J]. 新型电力系统, 2023(1): 32-43.
GUO J B, WANG T Z, LUO K, et al. Development of new power systems: challenges and solutions[J]. New type power systems, 2023(1): 32-43.
[2] ZHEN Z, LIU J M, ZHANG Z Y, et al.Deep learning based surface irradiance mapping model for solar PV power forecasting using sky image[J]. IEEE transactions on industry applications, 2020, 56(4): 3385-3396.
[3] 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 C, WANG Z, BAI J Y, et al. Review of ultra-short-term forecasting methods for photovoltaic power generation[J]. High voltage engineering, 2023, 49(7): 2938-2951.
[4] 龚莺飞, 鲁宗相, 乔颖, 等. 光伏功率预测技术[J]. 电力系统自动化, 2016, 40(4): 140-151.
GONG Y F, LU Z X, QIAO Y, et al. An overview of photovoltaic energy system output forecasting technology[J]. Automation of electric power systems, 2016, 40(4): 140-151.
[5] 朱想, 居蓉蓉, 程序, 等. 组合数值天气预报与地基云图的光伏超短期功率预测模型[J]. 电力系统自动化, 2015, 39(6): 4-10, 74.
ZHU X, JU R R, CHENG X, et al. A very short-term prediction model for photovoltaic power based on numerical weather prediction and ground-based cloud images[J]. Automation of electric power systems, 2015, 39(6): 4-10, 74.
[6] 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): 3027-3048.
ZHU Q F, LI J T, QIAO J, et al. Application and prospect of artificial intelligence technology in renewable energy forecasting[J]. Proceedings of the CSEE, 2023, 43(8): 3027-3048.
[7] SCOTT C, AHSAN M, ALBARBAR A.Machine learning for forecasting a photovoltaic (PV) generation system[J]. Energy, 2023, 278: 127807.
[8] 刘科研, 盛万兴, 马晓晨, 等. 基于多种群遗传算法的分布式光伏接入配电网规划研究[J]. 太阳能学报, 2021, 42(6): 146-155.
LIU K Y, SHENG W X, MA X C, et al. Planning research of distributed photovoltaic source access distribution network based on multi-population genetic algorithm[J]. Acta energiae solaris sinica, 2021, 42(6): 146-155.
[9] 吴春华, 董阿龙, 李智华, 等. 基于图相似日和PSO-XGBoost的光伏功率预测[J]. 高电压技术, 2022, 48(8): 3250-3259.
WU C H, DONG A L, LI Z H, et al. Photovoltaic power prediction based on graph similarity day and PSO-XGBoost[J]. High voltage engineering, 2022, 48(8): 3250-3259.
[10] LIU J X, ZANG H X, CHENG L L, et al.A Transformer-based multimodal-learning framework using sky images for ultra-short-term solar irradiance forecasting[J]. Applied energy, 2023, 342: 121160.
[11] SONG H, AL KHAFAF N, KAMOONA A, et al.Multitasking recurrent neural network for photovoltaic power generation prediction[J]. Energy reports, 2023, 9: 369-376.
[12] ZHANG C, PENG T, NAZIR M S.A novel integrated photovoltaic power forecasting model based on variational mode decomposition and CNN-BiGRU considering meteorological variables[J]. Electric power systems research, 2022, 213: 108796.
[13] LI Y, YE F, LIU Z H, et al.A short-term photovoltaic power generation forecast method based on LSTM[J]. Mathematical problems in engineering, 2021, 2021(1): 6613123.
[14] LIU J R, FU Y C.Decomposition spectral graph convolutional network based on multi-channel adaptive adjacency matrix for renewable energy prediction[J]. Energy, 2023, 284: 129242.
[15] 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.
[16] 王献志, 曾四鸣, 周雪青, 等. 基于XGBoost联合模型的光伏发电功率预测[J]. 太阳能学报, 2022, 43(4): 236-242.
WANG X Z, ZENG S M, ZHOU X Q, et al. Power forecast of photovoltaic generation based on XGBoost combined model[J]. Acta energiae solaris sinica, 2022, 43(4): 236-242.
[17] 谭海旺, 杨启亮, 邢建春, 等. 基于XGBoost-LSTM组合模型的光伏发电功率预测[J]. 太阳能学报, 2022, 43(8): 75-81.
TAN H W, YANG Q L, XING J C, et al. Photovoltaic power prediction based on combined XGBoost-LSTM model[J]. Acta energiae solaris sinica, 2022, 43(8): 75-81.
[18] LIU S K, JOHNS E, DAVISON A J.End-to-end multi-task learning with attention[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach, CA, USA, 2019: 1871-1880.
[19] 孟安波, 许炫淙, 陈嘉铭, 等. 基于强化学习和组合式深度学习模型的超短期光伏功率预测[J]. 电网技术, 2021, 45(12): 4721-4728.
MENG A B, XU X C, CHEN J M, et al. Ultra short term photovoltaic power prediction based on reinforcement learning and combined deep learning model[J]. Power system technology, 2021, 45(12): 4721-4728.
[20] WANG Y P, ZHAO J J, YANG C N, et al.Remaining useful life prediction of rolling bearings based on Pearson correlation-KPCA multi-feature fusion[J]. Measurement, 2022, 201: 111572.
[21] 彭曙蓉, 郑国栋, 黄士峻, 等. 基于XGBoost算法融合多特征短期光伏发电量预测[J]. 电测与仪表, 2020, 57(24): 76-83.
PENG S R, ZHENG G D, HUANG S J, et al. Multiple-feature short-term photovoltaic generation forecasting based on XGBoost algorithm[J]. Electrical measurement & instrumentation, 2020, 57(24): 76-83.
[22] GAO M M, LI J J, HONG F, et al.Day-ahead power forecasting in a large-scale photovoltaic plant based on weather classification using LSTM[J]. Energy, 2019, 187: 115838.
[23] GE H W, GAO D W, SUN L, et al.Multi-agent transfer reinforcement learning with multi-view encoder for adaptive traffic signal control[J]. IEEE transactions on intelligent transportation systems, 2022, 23(8): 12572-12587.