Data Recovery of Distributed Power Station Output Considering the Output Information of Correlated Stations
DOI:
https://doi.org/10.52152/4142Keywords:
Distributed Photovoltaic, Data Recovery, Neural Networks, Correlated Stations, Time-Series ImputationAbstract
The data acquisition devices of distributed photovoltaic power stations often lack proper maintenance, leading to frequent output data loss. This paper proposes a data recovery method that leverages the output information of correlated stations. First, the correlation between PV stations within the same region is calculated based on historical output data, and highly correlated station datasets are selected. Then, the complete data from these stations and the missing data from the target station are integrated and input into neural network models for recovery. Experimental results on the Desert Knowledge Solar Centre dataset show that incorporating correlated station data significantly improves accuracy. The TCN model achieves a 71.50% improvement, and the GRU model achieves 55.82%, outperforming other models due to their ability to capture temporal dependencies. This study's novelty lies in utilizing correlated station output instead of meteorological data, making it more practical for real-world PV data recovery.
References
R. Ahmed, V. Sreeram, Y. Mishra, M.D. Arif. A Review Evaluation of the State-of-the-Art in PV Solar Power Forecasting: Techniques Optimization. Renewable and Sustainable Energy Reviews, 2020, 124, 109792. DOI: 10.1016/j.rser.2020.109792
U.K. Das, K.S. Tey, M. Seyedmahmoudian, S. Mekhilef, M.Y.I. Idris, et al. Forecasting of Photovoltaic Power Generation Model Optimization: A Review. Renewable and Sustainable Energy Reviews, 2018, 91, 912-928. DOI: 10.1016/j.rser.2017.08.017
I. Pratama, A.E. Permanasari, I. Ardiyanto, R. Indrayani. A Review of Missing Values Handling Methods on Time-Series Data. 2016 International Conference on Information Technology Systems and Innovation (ICITSI), 2016, 1-6. DOI: 10.1109/ICITSI.2016.7858189
H. De Silva, A. Shehan Perera. Missing Data Imputation Using Evolutionary k-Nearest Neighbor Algorithm for Gene Expression Data. 2016 Sixteenth International Conference on Advances in ICT for Emerging Regions (ICTer), 2016, 141-146. DOI: 10.1109/ICTER.2016.7829911
F. Camastra, V. Capone, A. Ciaramella, A. Riccio, A. Staiano, et al. Prediction of Environmental Missing Data Time Series by Support Vector Machine Regression Correlation Dimension Estimation. Environmental Modelling & Software, 2022, 150, 105343. DOI: 10.1016/j.envsoft.2022.105343
F. Tang, H. Ishwaran. Random Forest Missing Data Algorithms. Statistical Analysis and Data Mining: An ASA Data Science Journal, 2017, 10(6), 363-377. DOI: 10.1002/sam.11348
V. Ravi, M. Krishna. A New Online Data Imputation Method Based on General Regression Auto Associative Neural Network. Neurocomputing, 2014, 138, 106-113. DOI: 10.1016/j.neucom.2014.02.037
Z.J. Sun, L. Xue, Y.M. Xu, Z. Wang. A Review of Deep Learning Research. Application Research of Computers, 2012, 29(08), 2806-2810. DOI: 10.3969/j.issn.1001-3695.2012.08.002
R.Y. Liao, Y.B. Liu, Y.D. Shen, H.J. Gao, D.L. Tang, et al. A Coupling Enhancement Method for Time-Series Data of Distributed Photovoltaic Clusters Based on Bidirectional Recurrent Interpolation Network. Power System Technology, 2024, 18(7), 2784-2794. DOI: 10.13335/j.1000-3673.pst.2-023.1681
W.J. Zhang, Y.H. Luo, Y. Zhang, D. Srinivasan. SolarGAN: Multivariate Solar Data Imputation Using Generative Adversarial Network. IEEE Transactions on Sustainable Energy, 2020, 12(1), 743-746. DOI: 10.1109/TSTE.2020.3004751
D.T. Neves, M.G. Naik, A. Proença. SGAIN, WSGAIN-CP, WSGAIN-GP: Novel GAN Methods for Missing Data Imputation. International Conference on Computational Science. Cham: Springer International Publishing, 2021, 98-113. DOI：10.1007/978-3-030-77961-0_10
K.Y. Liu, F.Z. Zhou, H. Zhou. Distribution Network Substation Data Restoration Based on Time-Series Signal Image Encoding Generative Adversarial Network. Power System Protection Control, 2022, 50(24), 129-136. DOI: 10.19783/j.cnki.pspc.2-20256
X.J. Qu, Z. Liu, C.Q. Wu, A.Q. Hou, X.Y. Yin, et al. MFGAN: Multimodal Fusion for Industrial Anomaly Detection Using Attention-Based Autoencoder Generative Adversarial Network. Sensors. 2024, 24(2), 637. DOI:10.3390/s24020637
E. Oh, T. Kim, Y.H. Ji, S. hyalia. STING: Self-Attention Based Time-Series Imputation Networks Using GAN. 2021 IEEE International Conference on Data Mining (ICDM), 2021, 1264-1269. DOI: 10.1109/ICDM51629.2021.00155
Y. Zhang, B.H. Zhou, X.R. Cai, W.Y. Guo, X.K. Ding, et al. Missing Value Imputation in Multivariate Time Series with End-to-End Generative Adversarial Networks. Information Sciences, 2021, 551, 67-82. DOI: 10.1016/j.-ins.2020.11.035
L.F. Xu, L.Y. Xu, J. Yu. Time Series Imputation with GAN Inversion Decay Connection. Information Sciences, 2023, 643, 119234. DOI: 10.1016/j.ins.202-3.119234
Q.T. Phan, Y.K. Wu, Q.D. Phan, H. Lo. A study on missing data imputation methods for improving hourly solar dataset. 2022 8th International Conference on Applied System Innovation(ICASI), 2022, 21-24. DOI: 10.1109/ICASI55125.2022.9774453
M. Shen, H.Z. Zhang, Y.X. Cao, F. Yang, Y.G. Wen. Missing data imputation for solar yield prediction using temporal multi-modal variational auto-encoder. Proceedings of the 29th ACM International Conference on Multimedia. 2021, 2558 - 2566. DOI: 10.1145/34740-85.3475430
X.Y. Zhang, M. Wen, J.X. Li, H.Y. Huang. Distributed PV Power Forecasting Considering Spatio-Temporal Dependence. International Conference on Smart Grid Smart Cities (ICSGSC), 2023, 487-493. DOI: 10.1109/ICSGSC59580.2023.10319128
S. Hochreiter, J. Schmidhuber. Long Short-Term Memory. Neural Computation, 1997, 9(8), 1735-1780. DOI: 10.1162/neco.1997.9.8.1735
J. Ma, J.C.P. Cheng, F.F. Jiang, W.W. Chen, M.Z. Wang, et al. A Bi-Directional Missing Data Imputation Scheme Based on LSTM Transfer Learning for Building Energy Data. Energy and Buildings, 2020, 216, 109941. DOI: 10.1016/j.enbuild.2020.109941
D. Li, L.H. Li, X.L. Li, Z.W. Ke, Q.H. Hu. Smoothed LSTM-AE: A Spatio-Temporal Deep Model for Multiple Time-Series Missing Imputation. Neurocomputing, 2020, 411, 351-363. DOI: 10.1016/j.ne-ucom.2020.05.033
N. Li, F.X. He, W.T. Ma, L. Jiang, X.P. Zhang. State of Charge Estimation of Lithium-Ion Batteries Based on Empirical Mode Decomposition Gated Recurrent Unit Neural Network. Transactions of China Electrotechnical Society, 2022, 37(17), 4528-4536. DOI: 10.19595/j.cnki.1000-6753.tces.211069
Z.P. Zhang, Y.Q. Zhang, A. Zeng, D. Pan, Y.Z. Ji, et al. Time-Series Data Imputation via Realistic Masking-Guided Tri-Attention Bi-GRU. ECAI, 2023, 3074-3082. DOI: 10.3233/FAIA230625
Q.X. Tan, M. Ye, B.Y. Yang, S.Q. Liu, A.J. Ma, et al. Data-GRU: Dual-Attention Time-Aware Gated Recurrent Unit for Irregular Multivariate Time Series. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(1), 930-937. DOI: 10.1609/aaai.v34i01.5440
Q.T. Li, Y. Xu. VS-GRU: A Variable Sensitive Gated Recurrent Neural Network for Multivariate Time Series with Massive Missing Values. Applied Sciences, 2019, 9(15), 3041. DOI: 10.3390/app9153041
L. Zhang, G.F. Ren, S.L. Li, J.S. Du, D.Y. Xu, et al. A novel soft sensor approach for industrial quality prediction based TCN with spatial temporal attention. Chemometrics Intelligent Laboratory Systems, 2025, 257, 105272. DOI: 10.1016/j.chemolab.2024.105272
Y.F. Mao, M. Yang, P. Li, Z.J. Ou. A Missing Data Imputation Method for Electricity Consumption Data Based on TCN-Attention with Mask Tokens. 2024 4th International Conference on Consumer Electronics Computer Engineering (ICCECE), 2024, 513-517. DOI: 10.1109/ICCECE61317.2024.10504227
K.K.R. Samal. Auto Imputation Enabled Deep Temporal Convolutional Network (TCN) Model for PM2.5 Forecasting. EAI Endorsed Transactions on Scalable Information Systems, 2025, 12(1). DOI: 10.4108/eetsis.5102
Z. Luo, Y.H. Wu, J.X. Zhu, W.J. Zhao, G. Wang, et al. Ultra-Short-Term Wind Power Forecasting Based on a Multi-Scale Time Series Block Autoencoder Transformer Neural Network Model. Power System Technology, 2023, 47(09), 3527-3537. DOI: 10.1109/ACCESS.2024.337-3798
J. Liu, S. Pasumarthi, B. Duffy, E. Gong, K. Datta, et al. One Model to Synthesize Them All: Multi-Contrast Multi-Scale Transformer for Missing Data Imputation. IEEE Transactions on Medical Imaging, 2023, 49(2), 2577-2591. DOI: 10.1109/TMI.2023.3261707
A. Lotfipoor, S. Patidar, D.P. Jenkins. Transformer Network for Data Imputation in Electricity DemData. Energy and Buildings, 2023, 300, 113675. DOI: 10.1016/j.enbuild.2023.113675
A. Yarkın Yıldız, E. Koç, A. Koç. Multivariate Time Series Imputation with Transformers. IEEE Signal Processing Letters, 2022, 29, 2517-2521. DOI: 10.1109/LSP.2022.3224880
J. Benesty, J.D. Chen, Y.T. Huang, I. Cohen. Pearson Correlation Coefficient. Noise Reduction in Speech Processing, 2009, 1-4. DOI: 10.1007/978-3-642-00296-0_5
J. Adler, I. Parmryd. Quantifying Colocalization by Correlation: The Pearson Correlation Coefficient is Superior to the Mander's Overlap Coefficient. Cytometry Part A, 2010, 77(8), 733-742. DOI: 1-0.1002/cyto.a.20896
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Haiyan Zeng, Xiangli Peng, Chenxi Dong, Ran Wu, Jinqiang Lin, Yaqi Wang (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
