SolarGAN: Multivariate Solar Data Imputation Using Generative Adversarial Network

被引：60

作者：

Zhang, Wenjie ^{[1
]}

Luo, Yonghong ^{[2
]}

Zhang, Ying ^{[2
]}

Srinivasan, Dipti ^{[1
]}

机构：

[1] Natl Univ Singapore NUS, Dept Elect & Comp Engn, Singapore 117576, Singapore

[2] Nankai Univ, Coll Comp Sci, Tianjin 300071, Peoples R China

来源：

IEEE TRANSACTIONS ON SUSTAINABLE ENERGY | 2021年 / 12卷 / 01期

关键词：

Gallium nitride; Generative adversarial networks; Training; Time series analysis; Forecasting; Indexes; Machine learning; Solar data imputation; generative adversarial network (GAN); PV forecasting; machine learning; smart grid;

D O I：

10.1109/TSTE.2020.3004751

中图分类号：

X [环境科学、安全科学];

学科分类号：

08 ; 0830 ;

摘要：

Photovoltaic (PV) is receiving increasing attention due to its sustainability and low carbon footprint. However, the penetration level of PV is still relatively low because of its intermittency. This uncertainty can be handled by accurate PV forecasting, which requires high-quality solar data. Nevertheless, up to 40% of solar data can be found missing, which significantly worsens the quality of solar data. This letter proposes a novel solarGAN method for multivariate solar data imputation, in which necessary modifications are made on the input of generative adversarial network (GAN) to effectively tackle the relatively independent solar time series data. Case studies on a public dataset show that the proposed solarGAN outperforms several commonly-used machine learning and GAN based data imputation methods with at least 23.9% reduction of mean squared error.

引用

页码：743 / 746

页数：4

共 50 条

[41] Generative Adversarial Network for Generating Different Types of Data
Murota S.
Iima H.
IEEJ Transactions on Electronics, Information and Systems, 2022, 142 (07) : 781 - 787
[42] Motor Fault Diagnosis Based on Generative Adversarial Network Using Hyperchaotic Sequences and Mixed-Dimensional Network
Li, Houzhen
Yao, Lina
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2025,
[43] Anonymization Through Data Synthesis Using Generative Adversarial Networks (ADS-GAN)
Yoon, Jinsung
Drumright, Lydia N.
van der Schaar, Mihaela
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2020, 24 (08) : 2378 - 2388
[44] Multi-Attention Generative Adversarial Network for Multivariate Time Series Prediction
Yin, Xiang
Han, Yanni
Sun, Hongyu
Xu, Zhen
Yu, Haibo
Duan, Xiaoyu
IEEE ACCESS, 2021, 9 : 57351 - 57363
[45] GENERATIVE ADVERSARIAL NETWORKS FOR GRAPH DATA IMPUTATION FROM SIGNED OBSERVATIONS
Madapu, Amarlingam
Segarra, Santiago
Chepuri, Sundeep Prabhakar
Marques, Antonio G.
2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 9085 - 9089
[46] Image Clustering Using an Augmented Generative Adversarial Network and Information Maximization
Ntelemis, Foivos
Jin, Yaochu
Thomas, Spencer A.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2022, 33 (12) : 7461 - 7474
[47] Unsupervised Multi-Discriminator Generative Adversarial Network for Lung Nodule Malignancy Classification
Kuang, Yan
Lan, Tian
Peng, Xueqiao
Selasi, Gati Elvis
Liu, Qiao
Zhang, Junyi
IEEE ACCESS, 2020, 8 (77725-77734) : 77725 - 77734
[48] Data augmentation through multivariate scenario forecasting in Data Centers using Generative Adversarial Networks
Perez, Jaime
Arroba, Patricia
Moya, Jose M.
APPLIED INTELLIGENCE, 2023, 53 (02) : 1469 - 1486
[49] Cropland Change Detection With Harmonic Function and Generative Adversarial Network
Chen, Jiage
Zhao, Wenzhi
Chen, Xi
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2022, 19
[50] A Conditional Generative adversarial Network for energy use in multiple buildings using scarce data
Baasch, Gaby
Rousseau, Guillaume
Evins, Ralph
ENERGY AND AI, 2021, 5

← 1 2 3 4 5 →