Combined IXGBoost-KELM short-term photovoltaic power prediction model based on multidimensional similar day clustering and dual decomposition

被引:17
作者
Wu, Thomas [1 ,2 ,3 ]
Hu, Ruifeng [1 ,3 ]
Zhu, Hongyu [1 ,2 ,3 ]
Jiang, Meihui [1 ,3 ]
Lv, Kunye [1 ,3 ]
Dong, Yunxuan [2 ,3 ]
Zhang, Dongdong [1 ,3 ]
机构
[1] Guangxi Univ, Sch Elect Engn, Nanning, Peoples R China
[2] Guangxi Univ, Sch Comp & Elect Informat, Nanning, Peoples R China
[3] Guangxi Univ, State Key Lab Featured Met Mat & Life Cycle Safety, Nanning, Peoples R China
基金
中国国家自然科学基金;
关键词
Photovoltaic prediction; Similar day clustering; Dual decomposition; Composite model; Golden jackal optimization; OPTIMIZATION; WIND; CONSUMPTION; GENERATION; NETWORK; DEMAND; OUTPUT;
D O I
10.1016/j.energy.2023.129770
中图分类号
O414.1 [热力学];
学科分类号
摘要
Accurate photovoltaic power prediction is important to ensure stable and safe operation of microgrids. However, due to the high volatility of photovoltaic power data, the prediction accuracy of traditional prediction models is often unsatisfactory. To ensure stable operation of microgrids, this study proposes a combined improved extreme gradient boosting-kernel extreme learning machine short-term photovoltaic power prediction model consisting of multidimensional similar day clustering and dual decomposition. Initially, gray relation analysis, Pearson correlation coefficient, and Kmeans++ are used for clustering to obtain high-precision similar days. Subsequently, a dual signal decomposition model based on variational modal decomposition and complete ensemble empirical mode decomposition with adaptive noise is proposed. Finally, predictions are made using a combination of predictive models with complementary strengths and weaknesses, and the prediction results of each component are fitted with under three weather conditions, the average root mean square error is reduced by 78.02%,62.99%, and 62.48%, and the average mean absolute error is reduced by 82.55%, 71.13%, and 67.07% in comparison with the baseline model. The results show that the model is effective in improving the prediction accuracy in a variety of different environments.
引用
收藏
页数:17
相关论文
共 50 条
[31]   Interval prediction of ultra-short-term photovoltaic power based on a hybrid model [J].
Zhang, Jinliang ;
Liu, Ziyi ;
Chen, Tao .
ELECTRIC POWER SYSTEMS RESEARCH, 2023, 216
[32]   Short-term photovoltaic power prediction model based on hierarchical clustering of K-means plus plus algorithm and deep learning hybrid model [J].
Wang, Man ;
Ma, Xiaojing ;
Wang, Ru ;
Kari, Tusongjiang ;
Tang, Zihan .
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY, 2024, 16 (02)
[33]   Short-Term Prediction of Photovoltaic Power Based on TCN-LSTM-Attention model and Kmeans plus [J].
Chen, Yuanjun ;
Wu, Jingtao ;
Wang, Yuee ;
Liang, Bin ;
Dong, Chunhui ;
He, Lian ;
Yu, Simiao .
2024 10TH INTERNATIONAL CONFERENCE ON POWER ELECTRONICS SYSTEMS AND APPLICATIONS, PESA 2024, 2024,
[34]   Short-term wind farm cluster power prediction based on dual feature extraction and quadratic decomposition aggregation [J].
Qu, Zhijian ;
Hou, Xinxing ;
Li, Jian ;
Hu, Wenbo .
ENERGY, 2024, 290
[35]   Short-term wind power prediction based on preprocessing and improved secondary decomposition [J].
Goh, Hui Hwang ;
He, Ronghui ;
Zhang, Dongdong ;
Liu, Hui ;
Dai, Wei ;
Lim, Chee Shen ;
Kurniawan, Tonni Agustiono ;
Teo, Kenneth Tze Kin ;
Goh, Kai Chen .
JOURNAL OF RENEWABLE AND SUSTAINABLE ENERGY, 2021, 13 (05)
[36]   Photovoltaic Power-Stealing Identification Method Based on Similar-Day Clustering and QRLSTM Interval Prediction [J].
Peng, Shurong ;
Guo, Lijuan ;
Li, Bin ;
Lu, Shuang ;
Chen, Huixia ;
Su, Sheng .
APPLIED SCIENCES-BASEL, 2023, 13 (06)
[37]   Short-Term Photovoltaic Power Generation Prediction Based on Copula Function and CNN-CosAttention-Transformer [J].
Hu, Keyong ;
Fu, Zheyi ;
Lang, Chunyuan ;
Li, Wenjuan ;
Tao, Qin ;
Wang, Ben .
SUSTAINABILITY, 2024, 16 (14)
[38]   Short-term forecasting of wind power generation Based on the Similar Day and Elman Neural Network [J].
Zhang, Xiaoyu ;
Wang, Rui ;
Liao, Tianjun ;
Zhang, Tao ;
Zha, Yabin .
2015 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI), 2015, :647-650
[39]   Hourly Day-Ahead Wind Power Prediction Using the Hybrid Model of Variational Model Decomposition and Long Short-Term Memory [J].
Shi, Xiaoyu ;
Lei, Xuewen ;
Huang, Qiang ;
Huang, Shengzhi ;
Ren, Kun ;
Hu, Yuanyuan .
ENERGIES, 2018, 11 (11)
[40]   Short-Term Power Forecasting Model for Photovoltaic Plants Based on Historical Similarity [J].
Monteiro, Claudio ;
Santos, Tiago ;
Alfredo Fernandez-Jimenez, L. ;
Ramirez-Rosado, Ignacio J. ;
Sonia Terreros-Olarte, M. .
ENERGIES, 2013, 6 (05) :2624-2643