An adaptive photovoltaic power interval prediction based on multi-objective optimization

被引:1
作者
Jiang, Yunxiao [1 ]
Wang, Xinyan [1 ]
Yang, Di [2 ]
Cheng, Runkun [2 ]
Zhao, Yinchuan [1 ]
Liu, Da [2 ]
机构
[1] North China Elect Power Univ, Sch Math & Phys, Beijing 102206, Peoples R China
[2] North China Elect Power Univ, Sch Econ & Management, Beijing 102206, Peoples R China
关键词
Photovoltaic prediction; Deep learning; Adaptive intervals; Multi-objective optimization; WIND;
D O I
10.1016/j.compeleceng.2024.109717
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Photovoltaic (PV) power interval prediction can provide a variation range of prediction results, which is of great significance to promoting the optimization of power grid dispatching and maintaining the stability of the power system. However, the existing PV interval prediction fails to accurately reflect error fluctuations in their construction process due to dependency on data distribution assumptions and unreasonable estimation of prediction error standard deviations (PEStds). To address these issues, we propose an adaptive interval prediction method based on multi-objective optimization. Firstly, K-means clustering is used to segment similar days. Subsequently, Convolutional Neural Network-Gated Recurrent Unit-Attention (CNN-GRU-Attention) is utilized for point prediction. Furthermore, Grey Relation Analysis (GRA) is used to screen dates similar to the target date under the same weather type, enabling a more precise estimation of prediction error fluctuations. Ultimately, the estimated values are multiplied by non-fixed parameters to construct intervals, which avoids the dependency on data distribution assumptions. Among them, the optimal values for non-fixed parameters are obtained through multi-objective optimization considering interval coverage probability, interval width, and deviation. To verify the effectiveness of the proposed model, we conduct comparative experiments on two datasets with different resolutions. The results demonstrate that the proposed model offers more flexible and higher-quality intervals. Not only does this research improve the accuracy of PV power generation interval prediction, but it also helps to promote the development of smart grid technology and improve the adaptive ability of power systems facing dynamic environments and complex data, which has an important impact on the future energy management and power market.
引用
收藏
页数:24
相关论文
共 50 条
[31]   A novel interval forecasting system based on multi-objective optimization and hybrid data reconstruct strategy [J].
Wang, Jianzhou ;
Zhou, Yilin ;
Jiang, He .
EXPERT SYSTEMS WITH APPLICATIONS, 2023, 217
[32]   A Novel Multi-Objective Optimization Model for Integrated Photovoltaic/Hydroelectric Power Generation Operation [J].
Ding, Shidong ;
Zeng, Pingliang ;
Xing, Hao ;
Yang, Jingqi ;
Zhou, Qinyong .
PROCEEDINGS OF THE 38TH CHINESE CONTROL CONFERENCE (CCC), 2019, :7328-7333
[33]   Multi-objective optimization prediction model for building parameters of photovoltaic windows based on NSGA II-BP [J].
Zhang, Jiran ;
Zhang, Lingling ;
Ren, Panpan ;
Hao, Wengang ;
Xu, Ao .
CASE STUDIES IN THERMAL ENGINEERING, 2024, 64
[34]   Probabilistic prediction-based multi-objective optimization approach for multi-energy virtual power plant [J].
Li, Gangqiang ;
Zhang, Rongquan ;
Bu, Siqi ;
Zhang, Junming ;
Gao, Jinfeng .
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS, 2024, 161
[35]   An Interval-Based Approach for Evolutionary Multi-Objective Optimization of Project Portfolios [J].
Balderas, Fausto ;
Fernandez, Eduardo ;
Gomez-Santillan, Claudia ;
Rungel-Valdez, Nelson ;
Cruz, Laura .
INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING, 2019, 18 (04) :1317-1358
[36]   Regional planning of solar photovoltaic technology based on LCA and multi-objective optimization [J].
Yuan, Jing ;
Xu, Xiaozhen ;
Huang, Beijia ;
Li, Zeqiu ;
Wang, Yuyue .
RESOURCES CONSERVATION AND RECYCLING, 2023, 195
[37]   A reinforcement learning-based multi-objective optimization in an interval and dynamic environment [J].
Xu, Yue ;
Song, Yuxuan ;
Pi, Dechang ;
Chen, Yang ;
Qin, Shuo ;
Zhang, Xiaoge ;
Yang, Shengxiang .
KNOWLEDGE-BASED SYSTEMS, 2023, 280
[38]   Prediction and multi-objective optimization of sieve tray hydrodynamic performance based on deep learning [J].
Yuan, Xing ;
Deng, Xuan ;
Wang, Kehan ;
Ding, Zhongwei ;
Zhao, Hongkang ;
Xue, Jiaxing ;
Li, Qunsheng .
CHEMICAL ENGINEERING SCIENCE, 2025, 309
[39]   Particle swarm optimization algorithms for interval multi-objective optimization problems [J].
Zhang, En-Ze ;
Wu, Yi-Fei ;
Chen, Qing-Wei .
Kongzhi yu Juece/Control and Decision, 2014, 29 (12) :2171-2176
[40]   Dynamic Multi-objective Optimization Algorithm Based on Reference Point Prediction [J].
Ding J.-L. ;
Yang C.-E. ;
Chen L.-P. ;
Chai T.-Y. .
Zidonghua Xuebao/Acta Automatica Sinica, 2017, 43 (02) :313-320