A Mamba–transformer-based framework for wideband electric energy metering error correction in renewable energy grid-connected systems
Journal of Renewable and Sustainable Energy · 01.11.2026 · 18 (6)
DOI: 10.1063/5.0362236 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Renewable and Sustainable Energy, online erschienen am 11.11.2026. Verfasst von Chong Li und 5 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben.
Abstract (Originalsprache)
The widespread deployment of photovoltaic generation and wind power has introduced complex wideband disturbances, including harmonics, interharmonics, and supraharmonics, which cause conventional electric energy meters to deviate from their calibrated accuracy classes. IEC 62053-21 specifies accuracy requirements for Class 1 active-energy meters; however, the present composite dataset exhibits uncorrected errors of several percent under wideband conditions, highlighting the need for operating-condition-dependent correction. To address the frequency-dependent measurement errors in renewable grid-connected systems, a Mamba-transformer wideband metering error correction network (MTWEC) is proposed. The framework integrates a selective state space branch and a band-biased attention branch to simultaneously capture long-term error evolution and spectral coupling characteristics. A learnable gating mechanism adaptively fuses the two representations, while physics-constrained optimization enforces spectral consistency and energy conservation. Experiments on a composite dataset comprising inverter test-bench measurements, hardware-in-the-loop data, field recordings, and simulation samples demonstrate that MTWEC achieves a mean absolute error of 2.47 × 10−3 p.u. and an R2 value of 0.991, demonstrating improved error-estimation performance under the investigated wideband operating conditions. A three-month field deployment at a 300 MW hybrid station reduces the mean settlement deviation from 1.43% to 0.28%, while inference latency remains 6.8 ms per window, below the 10.24 ms window-update interval. These results demonstrate the effectiveness and practical potential of the proposed method for online wideband electric energy metering error correction in renewable grid-connected systems.
Zitieren
APA
Li, C., Shen, H., Guo, J., Ma, H., Wang, Y., & Guo, R. (2026). A Mamba–transformer-based framework for wideband electric energy metering error correction in renewable energy grid-connected systems. Journal of Renewable and Sustainable Energy, 18(6). https://doi.org/10.1063/5.0362236
Vancouver
Li C, Shen H, Guo J, Ma H, Wang Y, Guo R. A Mamba–transformer-based framework for wideband electric energy metering error correction in renewable energy grid-connected systems. Journal of Renewable and Sustainable Energy. 2026;18(6). doi:10.1063/5.0362236
BibTeX
@article{Li2026,
author = {Li, Chong and Shen, Hongtao and Guo, Juchuan and Ma, Hongming and Wang, Yi and Guo, Rongkun},
title = {A Mamba–transformer-based framework for wideband electric energy metering error correction in renewable energy grid-connected systems},
journal = {Journal of Renewable and Sustainable Energy},
year = {2026},
volume = {18},
number = {6},
doi = {10.1063/5.0362236}
}
EUVIMED