Doubly Structured Data Synthesis for Time-Series Energy-Use Data

被引:0
作者
Kim, Jiwoo [1 ]
Lee, Changhoon [2 ]
Jeon, Jehoon [3 ]
Choi, Jungwoong [2 ]
Kim, Joseph H. T. [1 ,3 ]
机构
[1] Yonsei Univ, Dept Stat & Data Sci, 50 Yonsei Ro, Seoul 03722, South Korea
[2] Korea Smart Grid Inst, 3F,Samwoo Bldg,32 Nonhyeon ro 86 Gil, Seoul 06223, South Korea
[3] Yonsei Univ, Dept Appl Stat, 50 Yonsei Ro, Seoul 03722, South Korea
关键词
data augmentation; energy data; energy management; electronic energy use; data privacy; synthetic data;
D O I
10.3390/s24248033
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
As the demand for efficient energy management increases, the need for extensive, high-quality energy data becomes critical. However, privacy concerns and insufficient data volume pose significant challenges. To address these issues, data synthesis techniques are employed to augment and replace real data. This paper introduces Doubly Structured Data Synthesis (DS2), a novel method to tackle privacy concerns in time-series energy-use data. DS2 synthesizes rate changes to maintain longitudinal information and uses calibration techniques to preserve the cross-sectional mean structure at each time point. Numerical analyses reveal that DS2 surpasses existing methods, such as Conditional Tabular GAN (CTGAN) and Transformer-based Time-Series Generative Adversarial Network (TTS-GAN), in capturing both time-series and cross-sectional characteristics. We evaluated our proposed method using metrics for data similarity, utility, and privacy. The results indicate that DS2 effectively retains the underlying characteristics of real datasets while ensuring adequate privacy protection. DS2 is a valuable tool for sharing and utilizing energy data, significantly enhancing energy demand prediction and management.
引用
收藏
页数:16
相关论文
共 50 条
[41]   Automated Selection of Time Series Forecasting Models for Financial Accounting Data: Synthetic Data Application [J].
Strimaitis, Rokas ;
Ramanauskaite, Simona ;
Stefanovic, Pavel .
ELECTRONICS, 2025, 14 (07)
[42]   A Data-Level Augmentation Framework for Time Series Forecasting With Ambiguously Related Source Data [J].
Ye, Rui ;
Dai, Qun .
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2025, 37 (07) :3855-3868
[43]   An Empirical Study on Data Augmentation for Pixelwise Satellite Image Time-Series Classification and Cross-Year Adaptation [J].
Yuan, Yuan ;
Lin, Lei ;
Xin, Qi ;
Zhou, Zeng-Guang ;
Liu, Qingshan .
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2025, 18 :5172-5188
[44]   A Bayesian multivariate factor analysis model for causal inference using time-series observational data on mixed outcomes [J].
Samartsidis, Pantelis ;
Seaman, Shaun R. ;
Harrison, Abbie ;
Alexopoulos, Angelos ;
Hughes, Gareth J. ;
Rawlinson, Christopher ;
Anderson, Charlotte ;
Charlett, Andre ;
Oliver, Isabel ;
De Angelis, Daniela .
BIOSTATISTICS, 2023, 25 (03) :867-884
[45]   Importance of replication in analyzing time-series gene expression data: Corticosteroid dynamics and circadian patterns in rat liver [J].
Tung T Nguyen ;
Richard R Almon ;
Debra C DuBois ;
William J Jusko ;
Ioannis P Androulakis .
BMC Bioinformatics, 11
[46]   Using Time-Series Generative Adversarial Networks to Synthesize Sensing Data for Pest Incidence Forecasting on Sustainable Agriculture [J].
Tai, Chen-Yu ;
Wang, Wun-Jhe ;
Huang, Yueh-Min .
SUSTAINABILITY, 2023, 15 (10)
[47]   SynTiSeD - Synthetic Time Series Data Generator [J].
Meiser, Michael ;
Duppe, Benjamin ;
Zinnikus, Ingo .
2023 11TH WORKSHOP ON MODELLING AND SIMULATION OF CYBER-PHYSICAL ENERGY SYSTEMS, MSCPES, 2023,
[48]   Motif Alignment for Time Series Data Augmentation [J].
Bahri, Omar ;
Li, Peiyu ;
Boubrahimi, Soukaina Filali ;
Hamdi, Shah Muhammad .
BIG DATA ANALYTICS AND KNOWLEDGE DISCOVERY, DAWAK 2023, 2023, 14148 :42-48
[49]   Decision Support Model for Time Series Data Augmentation Method Selection [J].
Joubaud, Dorian ;
Kubler, Sylvain ;
Lourenco, Raoni ;
Cordy, Maxime ;
Le Traon, Yves .
IEEE ACCESS, 2024, 12 :196553-196566
[50]   Advancing IoT Data Utilization: Generating and Evaluating Synthetic Time Series Data [J].
Portase, Raluca-Laura ;
Dragotoniu, Corina-Madalina ;
Lemnaru, Camelia ;
Dinsoreanu, Mihaela ;
Potolea, Rodica .
2024 IEEE 20TH INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTER COMMUNICATION AND PROCESSING, ICCP 2024, 2024, :143-150