Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM

Qing Peng; Wenjun Pang; Yaodong Li; Xianfu Gong; Xian Zhang; Guibin Wang; Mahmoud Mohamed Sayed Mohamed Hemdan

doi:10.1109/ACPEE69242.2026.11524671

Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM

Qing Peng
, Wenjun Pang
, Yaodong Li
, Xianfu Gong
, Xian Zhang^*
, Guibin Wang
, Mahmoud Mohamed Sayed Mohamed Hemdan

^*Corresponding author for this work

School of Robotics and Advanced Manufacture, Harbin Institute of Technology Shenzhen
Shenzhen University
China Southern Power Grid
Cairo University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Driven by global carbon neutrality targets, the widespread integration of renewable energy and electric vehicles defines modern power systems. However, intermittent renewable generation and stochastic EV charging introduce severe bidirectional uncertainty, threatening grid stability. To address this, we propose WGAN-ConvLSTM, a novel hybrid spatiotemporal forecasting framework. The WGAN module overcomes data scarcity by generating high-fidelity synthetic samples that strictly align with real-world distributions. Subsequently, the ConvLSTM network leverages this augmented data to extract complex spatial correlations and temporal evolutionary dynamics, enabling high-precision, collaborative forecasting of renewable generation and EV charging loads.

Original language	English
Title of host publication	Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026
Editors	Tek-Tjing Lie, Ningyi Dai, Youbo Liu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1048-1052
Number of pages	5
ISBN (Electronic)	9798331560676
DOIs	https://doi.org/10.1109/ACPEE69242.2026.11524671
State	Published - 2026
Externally published	Yes
Event	11th Asia Conference on Power and Electrical Engineering, ACPEE 2026 - Macau, China Duration: 14 Apr 2026 → 17 Apr 2026

Publication series

Name	Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026

Conference

Conference	11th Asia Conference on Power and Electrical Engineering, ACPEE 2026
Country/Territory	China
City	Macau
Period	14/04/26 → 17/04/26

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Electric vehicle load
Renewable energy
Spatio-temporal forecasting
WGAN-ConvLSTM

Access to Document

10.1109/ACPEE69242.2026.11524671

Cite this

Peng, Q., Pang, W., Li, Y., Gong, X., Zhang, X., Wang, G., & Sayed Mohamed Hemdan, M. M. (2026). Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM. In T.-T. Lie, N. Dai, & Y. Liu (Eds.), Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026 (pp. 1048-1052). (Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ACPEE69242.2026.11524671

Peng, Qing ; Pang, Wenjun ; Li, Yaodong et al. / Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM. Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026. editor / Tek-Tjing Lie ; Ningyi Dai ; Youbo Liu. Institute of Electrical and Electronics Engineers Inc., 2026. pp. 1048-1052 (Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026).

@inproceedings{004a0895ee3346b3ab73d79e1c386829,

title = "Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM",

abstract = "Driven by global carbon neutrality targets, the widespread integration of renewable energy and electric vehicles defines modern power systems. However, intermittent renewable generation and stochastic EV charging introduce severe bidirectional uncertainty, threatening grid stability. To address this, we propose WGAN-ConvLSTM, a novel hybrid spatiotemporal forecasting framework. The WGAN module overcomes data scarcity by generating high-fidelity synthetic samples that strictly align with real-world distributions. Subsequently, the ConvLSTM network leverages this augmented data to extract complex spatial correlations and temporal evolutionary dynamics, enabling high-precision, collaborative forecasting of renewable generation and EV charging loads.",

keywords = "Electric vehicle load, Renewable energy, Spatio-temporal forecasting, WGAN-ConvLSTM",

author = "Qing Peng and Wenjun Pang and Yaodong Li and Xianfu Gong and Xian Zhang and Guibin Wang and \{Sayed Mohamed Hemdan\}, \{Mahmoud Mohamed\}",

note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026 ; Conference date: 14-04-2026 Through 17-04-2026",

year = "2026",

doi = "10.1109/ACPEE69242.2026.11524671",

language = "英语",

series = "Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1048--1052",

editor = "Tek-Tjing Lie and Ningyi Dai and Youbo Liu",

booktitle = "Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026",

address = "美国",

}

Peng, Q, Pang, W, Li, Y, Gong, X, Zhang, X, Wang, G & Sayed Mohamed Hemdan, MM 2026, Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM. in T-T Lie, N Dai & Y Liu (eds), Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026. Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026, Institute of Electrical and Electronics Engineers Inc., pp. 1048-1052, 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026, Macau, China, 14/04/26. https://doi.org/10.1109/ACPEE69242.2026.11524671

Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM. / Peng, Qing; Pang, Wenjun; Li, Yaodong et al.
Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026. ed. / Tek-Tjing Lie; Ningyi Dai; Youbo Liu. Institute of Electrical and Electronics Engineers Inc., 2026. p. 1048-1052 (Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM

AU - Peng, Qing

AU - Pang, Wenjun

AU - Li, Yaodong

AU - Gong, Xianfu

AU - Zhang, Xian

AU - Wang, Guibin

AU - Sayed Mohamed Hemdan, Mahmoud Mohamed

PY - 2026

Y1 - 2026

N2 - Driven by global carbon neutrality targets, the widespread integration of renewable energy and electric vehicles defines modern power systems. However, intermittent renewable generation and stochastic EV charging introduce severe bidirectional uncertainty, threatening grid stability. To address this, we propose WGAN-ConvLSTM, a novel hybrid spatiotemporal forecasting framework. The WGAN module overcomes data scarcity by generating high-fidelity synthetic samples that strictly align with real-world distributions. Subsequently, the ConvLSTM network leverages this augmented data to extract complex spatial correlations and temporal evolutionary dynamics, enabling high-precision, collaborative forecasting of renewable generation and EV charging loads.

AB - Driven by global carbon neutrality targets, the widespread integration of renewable energy and electric vehicles defines modern power systems. However, intermittent renewable generation and stochastic EV charging introduce severe bidirectional uncertainty, threatening grid stability. To address this, we propose WGAN-ConvLSTM, a novel hybrid spatiotemporal forecasting framework. The WGAN module overcomes data scarcity by generating high-fidelity synthetic samples that strictly align with real-world distributions. Subsequently, the ConvLSTM network leverages this augmented data to extract complex spatial correlations and temporal evolutionary dynamics, enabling high-precision, collaborative forecasting of renewable generation and EV charging loads.

KW - Electric vehicle load

KW - Renewable energy

KW - Spatio-temporal forecasting

KW - WGAN-ConvLSTM

UR - https://www.scopus.com/pages/publications/105041631671

U2 - 10.1109/ACPEE69242.2026.11524671

DO - 10.1109/ACPEE69242.2026.11524671

M3 - 会议稿件

AN - SCOPUS:105041631671

T3 - Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026

SP - 1048

EP - 1052

BT - Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026

A2 - Lie, Tek-Tjing

A2 - Dai, Ningyi

A2 - Liu, Youbo

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026

Y2 - 14 April 2026 through 17 April 2026

ER -

Peng Q, Pang W, Li Y, Gong X, Zhang X, Wang G et al. Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM. In Lie TT, Dai N, Liu Y, editors, Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026. Institute of Electrical and Electronics Engineers Inc. 2026. p. 1048-1052. (Proceedings - 2026 11th Asia Conference on Power and Electrical Engineering, ACPEE 2026). doi: 10.1109/ACPEE69242.2026.11524671

Spatial-Temporal Data Augmentation and Prediction in Power Systems Based on WGAN-ConvLSTM

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this