A Comparative Study on Data Augmentation Using Generative Models for Robust Solar Irradiance Prediction

Jinyeong Oh (오진영); Jimin Lee (이지민); Daesungjin Kim (김대성진); Bo-Young Kim (김보영); Jihoon Moon (문지훈)

doi:10.9708/jksci.2023.28.11.029

A Comparative Study on Data Augmentation Using Generative Models for Robust Solar Irradiance Prediction

Journal of The Korea Society of Computer and Information
Abbr : JKSCI
2023, 28(11), pp.29-42
DOI : 10.9708/jksci.2023.28.11.029
Publisher : The Korean Society Of Computer And Information
Research Area : Engineering > Computer Science
Received : October 18, 2023
Accepted : November 13, 2023
Published : November 30, 2023

Jinyeong Oh ¹, Jimin Lee ¹, Daesungjin Kim ², Bo-Young Kim ², Jihoon Moon ¹

¹순천향대학교
²아산중학교

Accredited

ABSTRACT

In this paper, we propose a method to enhance the prediction accuracy of solar irradiance for three major South Korean cities: Seoul, Busan, and Incheon. Our method entails the development of five generative models—vanilla GAN, CTGAN, Copula GAN, WGANGP, and TVAE—to generate independent variables that mimic the patterns of existing training data. To mitigate the bias in model training, we derive values for the dependent variables using random forests and deep neural networks, enriching the training datasets. These datasets are integrated with existing data to form comprehensive solar irradiance prediction models. The experimentation revealed that the augmented datasets led to significantly improved model performance compared to those trained solely on the original data. Specifically, CTGAN showed outstanding results due to its sophisticated mechanism for handling the intricacies of multivariate data relationships, ensuring that the generated data are diverse and closely aligned with the real-world variability of solar irradiance. The proposed method is expected to address the issue of data scarcity by augmenting the training data with high-quality synthetic data, thereby contributing to the operation of solar power systems for sustainable development.

KEYWORDS

Solar Energy Forecasting, Generative Data Augmentation, Deep Learning Models, Environmental Sustainability, Data Insufficiency Solutions

Citation status

* References for papers published after 2023 are currently being built.

[journal] 한유나 / 2022 / A Robust and Explainable Missing Value Imputation Method for Smart Meter Data / 한국전자거래학회지 / 한국전자거래학회 27(3) : 21~43

[other] L. Cozzi / 2020 / World energy outlook 2020 / International Energy Agency : 1~461

[web] / 2023 SEMI-ANNAL Report / https://www.ctis.re.kr/ko/downloadBbsFile.do?atchmnflNo=9673

[journal] Abdeen Mustafa Omer / 2008 / Energy, environment and sustainable development / Renewable and Sustainable Energy Reviews / Elsevier BV 12(9) : 2265~2300

[journal] 문지훈 / 2022 / Multistep-Ahead Solar Irradiance Forecasting for Smart Cities Based on LSTM, Bi-LSTM, and GRU Neural Networks / 한국전자거래학회지 / 한국전자거래학회 27(4) : 27~52

[journal] Can Wan / 2015 / Photovoltaic and solar power forecasting for smart grid energy management / CSEE Journal of Power and Energy Systems / Power System Technology Press 1(4) : 38~46

[journal] Jinwoong Park / 2020 / Multistep-Ahead Solar Radiation Forecasting Scheme Based on the Light Gradient Boosting Machine: A Case Study of Jeju Island / Remote Sensing / MDPI AG 12(14) : 2271~

[journal] Dayeong So / 2023 / A Hybrid Ensemble Model for Solar Irradiance Forecasting: Advancing Digital Models for Smart Island Realization / Electronics / MDPI AG 12(12) : 2607~

[journal] Cyril Voyant / 2017 / Machine learning methods for solar radiation forecasting: A review / Renewable Energy / Elsevier BV 105 : 569~582

[journal] Dayeong So / 2023 / BiGTA-Net: A Hybrid Deep Learning-Based Electrical Energy Forecasting Model for Building Energy Management Systems / Systems / MDPI AG 11(9) : 456~

[journal] J. Lee / 2020 / Reliable solar irradiance prediction using ensemble learning-based models: A comparative study / Energy Conversion and Management 208 : 112582~

[journal] Pratima Kumari / 2021 / Deep learning models for solar irradiance forecasting: A comprehensive review / Journal of Cleaner Production / Elsevier BV 318 : 128566~

[journal] Ke Yan / 2020 / Short-Term Solar Irradiance Forecasting Based on a Hybrid Deep Learning Methodology / Information / MDPI AG 11(1) : 32~

[journal] Subeen Leem / 2023 / Enhancing multistep-ahead bike-sharing demand prediction with a two-stage online learning-based time-series model: insight from Seoul / The Journal of Supercomputing / Springer Science and Business Media LLC

[journal] Jie Gui / 2023 / A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications / IEEE Transactions on Knowledge and Data Engineering / Institute of Electrical and Electronics Engineers (IEEE) 35(4) : 3313~3332

[journal] Xiaoqiao Huang / 2022 / Time series forecasting for hourly photovoltaic power using conditional generative adversarial network and Bi-LSTM / Energy / Elsevier BV 246 : 123403~

[journal] Qianqian Li / 2023 / A Solar Irradiance Forecasting Framework Based on the CEE-WGAN-LSTM Model / Sensors / MDPI AG 23(5) : 2799~

[journal] Jaeuk Moon / 2020 / Conditional Tabular GAN-Based Two-Stage Data Generation Scheme for Short-Term Load Forecasting / IEEE Access / Institute of Electrical and Electronics Engineers (IEEE) 8 : 205327~205339

[journal] Fei Wang / 2019 / Generative adversarial networks and convolutional neural networks based weather classification model for day ahead short-term photovoltaic power forecasting / Energy Conversion and Management / Elsevier BV 181 : 443~462

[journal] Mohammad Navid Fekri / 2019 / Generating Energy Data for Machine Learning with Recurrent Generative Adversarial Networks / Energies / MDPI AG 13(1) : 130~

[journal] Chenlu Tian / 2019 / Data driven parallel prediction of building energy consumption using generative adversarial nets / Energy and Buildings / Elsevier BV 186 : 230~243

[journal] L. Xu / 2019 / Modeling tabular data using conditional GAN / Advances in Neural Information Processing Systems 32

[web] S. Kamthe / 2021 / Copula flows for synthetic data generation / arXiv / 10.48550/arXiv.2101.00598

[confproc] M. Arjovsky / 2017 / Wasserstein Generative Adversarial Networks / Proceedings of the 34th International Conference on Machine Learning

[confproc] H. Ishfaq / 2018 / TVAE: Deep metric learning approach for variational autoencoder / roceedings of the Workshop at International Conference on Learning Representations 32

[confproc] J. Moon / 2021 / A comparative analysis of tree-based models for day-ahead solar irradiance forecasting / Proceedings of the 2021 International Conference on Platform Technology and Service : 1~6

[journal] Leo Breiman / 2001 / Random forests / Machine Learning / Springer Science and Business Media LLC 45(1) : 5~32

[journal] Joohyun Jang / 2023 / RAID: Robust and Interpretable Daily Peak Load Forecasting via Multiple Deep Neural Networks and Shapley Values / Sustainability / MDPI AG 15(8) : 6951~

[confproc] V. Borisov / 2022 / Deep neural networks and tabular data: A survey / IEEE Transactions on Neural Networks for Learning Systems : 1~21

[journal] Jihoon Moon / 2018 / Hybrid Short-Term Load Forecasting Scheme Using Random Forest and Multilayer Perceptron / Energies / MDPI AG 11(12) : 3283~

[web] / Orange Data Mining / https://orangedatamining.com

This paper was written with support from the National Research Foundation of Korea.

KJCKorea
Journal Central

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

A Comparative Study on Data Augmentation Using Generative Models for Robust Solar Irradiance Prediction

ABSTRACT

KEYWORDS

Citation status

* References for papers published after 2023 are currently being built.

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

A Comparative Study on Data Augmentation Using Generative Models for Robust Solar Irradiance Prediction

ABSTRACT

KEYWORDS

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (0)

REFERENCES (31) * References for papers published after 2023 are currently being built.

Search PDF

Citation

* References for papers published after 2023 are currently being built.