EUVIMEDScopri EUVIMED
Open access · CC BY 4.0

Enhancing Long-Term Forecasting Stability in Smart Grids: A Hybrid Mamba-LSTM-Attention Framework

Fusheng Chen, Chong Fo Lei, Te Guo, Chiawei Chu

Energies · 2026-04-09 · 19 (8) · 1855

DOI: 10.3390/en19081855 ↗

At a glance

Journal article in Energies, published online on 2026-04-09. Written by Fusheng Chen and 3 co-authors. Cited 1 times so far (Crossref); 46 references. The full text is freely available (open access, CC BY 4.0). Funded by: Shenzhen Great Bay Technology Joint Laboratory of Artificial Intelligence and Big Data.

AbstractDati bibliograficiCiteRelated studies

Abstract (original language)

Accurate multivariate long-term time series forecasting (LTSF) is critical for smart grid operations. However, non-stationary distribution shifts frequently induce compounding error accumulation in conventional architectures. This study proposes the Mamba-LSTM-Attention (MLA) framework, a distribution-aware architecture engineered for forecasting stability. The pipeline integrates Reversible Instance Normalization (RevIN) to neutralize statistical drift. To address computational bottlenecks, the architecture utilizes a linear-time Selective State Space Model (Mamba) to capture global trend dynamics, cascaded with a single-layer gated Long Short-Term Memory (LSTM) unit to model localized non-linear residuals. A terminal information bottleneck structurally bounds cross-step error propagation. Empirical results across standard ETT and Electricity benchmarks reveal a precision–stability trade-off. By prioritizing structural resilience, the MLA framework limits error accumulation on highly volatile datasets, yielding MSEs of 0.210 and 0.128 on ETTh2 and ETTm2 at the T = 96 horizon. This structural bottleneck inherently smooths high-frequency periodic patterns, yielding lower absolute accuracy on stationary benchmarks such as ETTh1 and ETTm1. Ultimately, the architecture establishes a computationally efficient, structurally stable baseline tailored for non-stationary anomaly tracking in smart grids.

Funding

  • Shenzhen Great Bay Technology Joint Laboratory of Artificial Intelligence and Big Data

Cite

APA

Chen, F., Lei, C. F., Guo, T., & Chu, C. (2026). Enhancing Long-Term Forecasting Stability in Smart Grids: A Hybrid Mamba-LSTM-Attention Framework. Energies, 19(8), 1855. https://doi.org/10.3390/en19081855

Vancouver

Chen F, Lei CF, Guo T, Chu C. Enhancing Long-Term Forecasting Stability in Smart Grids: A Hybrid Mamba-LSTM-Attention Framework. Energies. 2026;19(8):1855. doi:10.3390/en19081855

BibTeX
@article{Chen2026,
  author = {Chen, Fusheng and Lei, Chong Fo and Guo, Te and Chu, Chiawei},
  title = {Enhancing Long-Term Forecasting Stability in Smart Grids: A Hybrid Mamba-LSTM-Attention Framework},
  journal = {Energies},
  year = {2026},
  volume = {19},
  number = {8},
  pages = {1855},
  doi = {10.3390/en19081855}
}
Start for free →