Huang, J., Lee, G. C. F., Kurniawan, E., Boonkajay, A., & Sun, S. (2024). High-Reliability Wireless Communications for Virtual Power Plant with Efficient Network Redundancy. 2024 IEEE Globecom Workshops (GC Wkshps), 1–6. https://doi.org/10.1109/gcwkshp64532.2024.11100725
Abstract:
In contrast to traditional costly wired communication systems, high-reliability, low-latency wireless communication offers a more scalable and cost-efficient solution for transmitting information in virtual power plant (VPP) platforms to manage distributed energy resources (DERs).
In this paper, we introduce a pre-emptive on-demand redundant transmission strategy that leverages deep learning-based extreme event prediction to ensure low latency and resilient wireless communication for VPP applications. Accurate prediction of extreme events is critical for making energy-efficient redundant transmission decisions. However, performance degradation in prediction algorithms can occur due to the distribution shift between training and testing data. To address this, we propose an Adaptive Decomposition-Enhanced TimesNet (AD-TimesNet) that improves the prediction performance under such distribution shifts. The DER user equipment (UE) triggers transmissions on the backup network when extreme events are anticipated on the primary wireless link. We build a 5G-based testbed for data collection and evaluation. The results demonstrate that the proposed algorithm has the ability to meet low-latency and high-reliability targets for future-proof wireless VPP communication while significantly reducing transmission costs.
License type:
Publisher Copyright
Funding Info:
This work was supported by the National Research Foundation (NRF) Singapore, through Future Proof Reliable and Resilient Wireless Communications for Virtual Power Plant (W-VPP), under its Industry Alignment Fund (Prepositioning) (IAF-PP) for Urban Solutions and Sustainability (USS) Domain, Research Innovation Enterprise 2020 Plan (RIE2020), and Energy Grid 2.0 Programme.