Fourier-enhanced sequence-to-sequence latent graph neural networks for multi-node spatiotemporal forecasting in a hydroelectric reservoir

Seman, Laio OrielStefenon, Stefano FrizzoYow, Kin-ChoongCoelho, Leandro dos SantosMariani, Viviana Cocco2026-01-282026-01-282026-03-01Seman, L. O., Stefenon, S. F., Yow, K. C., Coelho, L. S., & Mariani, V. C. (2026). Fourier-enhanced sequence latent graph neural networks for multi-node spatiotemporal forecasting in a hydroelectric reservoir. Engineering Applications of Artificial Intelligence, 167, Part 3, 1-28. https://doi.org/10.1016/j.engappai.2026.1139390952-1976http://hdl.handle.net/10400.21/22584This work was supported in part by the Natural Sciences and Engineering Research Council of Canada, Canada (NSERC) under Grant DDG-2024-00035, and in part by the Cette recherche a été financée par le Conseil de recherches en sciences naturelles et en génie du Canada (CRSNG) under Grant DDG-2024-00035. The authors Mariani and Coelho thank the National Council of Scientific and Technological Development of Brazil - CNPq (Grants number: 314389/2023-7-PQ, 313169/2023-3-PQ, 407453/2023-7-Universal, and 442176/2023-6-Peci) for its financial support of this work.This paper presents a Fourier-enhanced dynamic sequence-to-sequence latent graph neural network (Seq2SeqLatentGNN), a deep learning architecture for multi-node spatiotemporal forecasting in hydroelectric reservoir systems. The model integrates three key components: (i) a custom Fourier layer that analyzes global temporal patterns through frequency-domain transformations, (ii) a latent correlation graph convolutional network that infers relational structures between monitoring stations without requiring predefined adjacency matrices, and (iii) an attention-based sequence-to-sequence model that processes temporal dependencies while enabling multi-step forecasting. The architecture simultaneously learns graph structure and forecasting tasks, adapting to changing spatial relationships between reservoir nodes. The proposed architecture was evaluated using a comprehensive dataset derived from 19 interconnected hydroelectric reservoirs located in southern Brazil. The dataset encompasses multiple years of high-resolution (hourly) measurements, including reservoir water levels, inflow and outflow rates, precipitation records, and energy production metrics. Experimental results demonstrate that Seq2SeqLatentGNN achieves superior performance compared to conventional statistical models and contemporary machine learning methods, as measured by standard error metrics. Analysis of the learned latent correlations reveals meaningful spatial dependencies that align with hydrological principles. The model exhibits consistent performance across varying temporal patterns, adapts to regime transitions, and captures both periodic and nonstationary dynamics. The proposed architecture contributes to spatiotemporal forecasting by combining spectral processing, dynamic graph learning, and sequence modeling in a unified framework applicable to systems with evolving connectivity patterns.engGraph neural networksFourier neural operatorSpatiotemporal forecastingHydroelectric reservoirsFourier-enhanced sequence-to-sequence latent graph neural networks for multi-node spatiotemporal forecasting in a hydroelectric reservoirjournal article10.1016/j.engappai.2026.113939