| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 3 MB | Adobe PDF |
Orientador(es)
Resumo(s)
This study presents a unified framework for thermal energy forecasting that integrates signal decomposition, Neural Architecture Search (NAS), and sparse transformer modeling. The method employs a hyperparameter-optimized hierarchical Variational Mode Decomposition (VMD) to enhance signal representation through adaptive noise suppression and mode separation. A differentiable preprocessing module, referred to as HeadComposer, performs joint search over normalization, decomposition, and convolutional transformations within a bilevel optimization procedure. The forecasting backbone consists of a transformer architecture incorporating gated residual connections for adaptive token selection and Mixture-of-Experts (MoE) feedforward layers for conditional computation. Experiments are conducted on real-world thermal power generation data comprising approximately 10,000 test samples, using an input look-back window of 96 time steps and a multi-step prediction horizon of 24 steps ahead. Results demonstrate consistent improvements over state-of-the-art baselines, achieving a mean absolute percentage error of 5.94%, corresponding to reductions of approximately 12-15% relative to competing methods. The proposed design demonstrates the effectiveness of combining decomposition-based preprocessing (VMD and HeadComposer) with NAS-driven architecture adaptation and sparse expert routing (based on MoE) for modeling non-stationary thermal generation processes.
Descrição
Palavras-chave
Mixture-of-experts Neural architecture search Variational mode decomposition
Contexto Educativo
Citação
Seman, L. O., Stefenon, S. F., Matos-Carvalho, J. P., Nied, A., & Yow, K.-C. (2026). End-to-End Decomposition-Aware Neural Architecture Search With Sparse Expert Transformers for Thermal Energy Forecasting. IEEE Access, 14, 67765–67787. https://doi.org/10.1109/ACCESS.2026.3689436
Editora
IEEE