Percorrer por autor "Seman, Laio Oriel"
A mostrar 1 - 10 de 11
Resultados por página
Opções de ordenação
- Decomposition-driven Mamba state space models with expert routing for wind curtailment forecastingPublication . Seman, Laio Oriel; Stefenon, Stefano Frizzo; Matos-Carvalho, João Pedro; Stefenon, Stefano Frizzo; ElsevierWind power curtailment, known in Brazil as constrained-off, has intensified with the rapid expansion of renew able generation, driven by transmission operational constraints. Accurate forecasting of curtailment is essential for grid planning, compensation mechanisms, and infrastructure optimization. However, constrained-off time series exhibit strong non-stationarity, multi-scale behavior, and long-range dependencies that challenge conventional forecasting models. This paper proposes a decomposition-driven Mamba state space framework with Mixture-of-Experts (MoE) routing for wind curtailment forecasting. The approach integrates Entropy-Guided Variational Mode Decomposition (EG-VMD) for adaptive trend and noise separation, Reversible Instance Normalization to mitigate distribution shift, multi-scale seasonal-trend modeling, and a sparse expert-based Mamba backbone with linear-time complexity. In benchmark comparisons against state-of-the-art forecasting models, the proposed MoE-Mamba reduces Mean Squared Error (MSE) by up to 27%, Root Mean Squared Error by 14.6%, and Mean Absolute Error by 16.1% relative to the strongest baselines, while improving upon a single-expert Mamba by 15.6% in MSE. The results highlight the effectiveness of decomposition-enhanced State Space Models with expert routing for scalable, long-horizon forecasting in renewable power systems.
- Differentiable neural search architecture with zero-cost metrics for insulator fault predictionPublication . Seman, Laio Oriel; Buratto, William Gouvêa; Gonzalez, Gabriel Villarrubia; Leithardt, Valderi Reis Quietinho; Nied, Ademir; Stefenon, Stefano FrizzoReliable monitoring of high-voltage insulators is critical for maintaining the stability of electrical power systems, particularly under environmental contamination that can lead to flashover. Traditional inspection techniques struggle to anticipate degradation dynamics, while data-driven models often rely on fixed neural architectures that inadequately capture the complex temporal patterns in leakage current signals. This work proposes a Differentiable Neural Architecture Search (DARTS) framework, based on zero-cost metrics, tailored for time series forecasting in insulator monitoring. The method based on DARTS integrates a mixed encoder-decoder design with learnable selection over long short-term memory, gated recurrent units, and transformer components, coupled with a cross-attention bridge featuring temporal bias and gating mechanisms. To ensure efficient architecture exploration, the search leverages metrics such as SynFlow and Jacobian covariance for early candidate screening, followed by a bilevel optimization stage with entropy and diversity regularization. Experiments on real-world leakage current data demonstrate that the discovered architectures outperform manually designed baselines, offering improved forecasting performance.
- End-to-end decomposition-aware neural architecture search with sparse expert transformers for thermal energy forecastingPublication . Seman, Laio Oriel; Stefenon, Stefano Frizzo; Matos-Carvalho, João Pedro; Nied, Ademir; Yow, Kin-Choong; Stefenon, Stefano Frizzo; IEEEThis 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.
- Enhanced random vector functional link networks with bayesian-based hyperparameter optimization for wind speed forecastingPublication . Seman, Laio Oriel ; Klaar, Anne Carolina Rodrigues ; Ribeiro, Matheus Henrique Dal Molin ; Stefenon, Stefano FrizzoAccurate short-term wind speed forecasting is essential for reliable and efficient wind energy integration. This paper introduces an enhanced Random Vector Functional Link (RVFL) network optimized through a Bayesian-based Neural Architecture Search (NAS) framework. The proposed RVFL-OptBayes model incorporates multi-scale feature generation, including kernel approximations, Nystr & ouml;m sampling, Fastfood transforms, wavelet scattering, and Neural Tangent Kernel embeddings with Principal Component Analysis (PCA)-aligned orthogonal initializations and spectral normalization to improve stability and feature diversity. Experiments were conducted on real-world Brazilian wind farm data to evaluate forecasting performance. Results show that RVFL-OptBayes outperforms conventional RVFL networks, deep learning models, and ensemble methods, achieving an R2 above 0.99. The proposed framework demonstrates that lightweight randomized architectures, when combined with principled hyperparameter search, can rival or surpass complex deep learning models for time-series forecasting. The findings suggest strong potential for practical deployment in renewable energy systems, offering accurate and computationally efficient wind speed predictions to support operational planning, grid stability, and smart energy management.
- Fourier-enhanced sequence-to-sequence latent graph neural networks for multi-node spatiotemporal forecasting in a hydroelectric reservoirPublication . Seman, Laio Oriel; Stefenon, Stefano Frizzo; Yow, Kin-Choong; Coelho, Leandro dos Santos; Mariani, Viviana CoccoThis 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.
- Input attention, squeeze and excitation, and spatial transformer of YOLO for fault detection using UAVPublication . Carvalho, João Pedro Matos; Stefenon, Stefano Frizzo; Leithardt, Valderi Reis Quietinho; Seman, Laio Oriel; Yow, Kin-Choong; Santana, Juan Francisco De PazThe detection of faults in insulators is important to guarantee the continuous supply of electricity. To identify faults in these components, various object detection methods based on deep learning have been explored. This paper investigates architectural enhancements to the You Only Look Once (YOLO) framework for fault detection in electrical power grid insulators. Three structural variants are proposed: the Input Attention Transformer (IAT-YOLO) for spatial feature refinement, Squeeze-and-Excitation (SAE-YOLO) modules for channel recalibration, and Spatial Transformer Networks (STN-YOLO) for geometric alignment. Experiments were conducted on a publicly available insulator dataset from Unmanned Aerial Vehicles (UAVs), comprising seven defect categories, including pollution, breakage, and flashover damage. Results demonstrate that STN-YOLO and SAE-YOLO consistently improve generalization and robustness, achieving mAP values of up to 0.995 for specific classes. The findings highlight the effectiveness of integrating attention mechanisms and spatial transformations to enhance YOLO-based detection, contributing to improved automated inspection of the power grid.
- Multi-step short-term solar energy forecasting using Fourier-enhanced BiLSTM and neural additive modelsPublication . Seman, Laio Oriel; Stefenon, Stefano Frizzo; Yow, Kin-Choong; Coelho, Leandro dos Santos; Mariani, Viviana CoccoAccurate short and medium-term forecasting is important for mitigating uncertainty and enabling efficient energy grid management. While traditional machine learning and deep learning models offer improved accuracy, they often lack interpretability. To address these limitations, this study proposes a hybrid forecasting framework, called FNO-BiLSTM-NAM, that combines a Fourier Neural Operator (FNO) to extract spectral–temporal features, a Bidirectional Long Short-Term Memory (BiLSTM) network to model sequential dependencies, and a Neural Additive Model (NAM) to quantify feature-wise contributions. The model incorporates multi-scenario forecasting to support energy operators under different uncertainty levels. Experiments conducted on a dataset from a 5 MW PhotoVoltaic (PV) plant demonstrate the superiority of the model. For a 6-hour forecast horizon, the proposed FNO-BiLSTM-NAM model achieved a mean absolute error of 0.0712 and mean squared error of 0.0092, outperforming benchmark models across short- to medium-term horizons. Furthermore, the spectral analysis of the FNO revealed low-pass filtering behavior, highlighting the ability of the model to suppress high-frequency noise. Comparative experiments with five machine and deep learning baseline models confirm the robustness and generalization capacity of the framework. These results underscore the potential of the proposed model for enhancing PV energy forecasting accuracy while maintaining transparency across dynamic operating conditions.
- Optimized Harmonic-Multiscale Kolmogorov–Arnold Network denoised by EWT for hydroelectric dam useful volume forecastPublication . Stefenon, Stefano Frizzo; Seman, Laio Oriel; Matos-Carvalho, João Pedro; Nied, Ademir; Gonzalez, Gabriel Villarrubia; Yow, Kin-Choong; Stefenon, Stefano Frizzo; ElsevierAccurate forecasting of the useful volume of a hydroelectric reservoir is essential for energy planning and water resource management. To address this challenge, this paper proposes a novel Harmonic-Multiscale Kolmogorov–Arnold Network forecasting model. The model integrates sinusoidal (Sin) basis functions to model global oscillatory patterns and wavelet (Wav) bases to capture localized transient features within the Kolmogorov Arnold Network (KAN) framework (SinWavKAN). This hybrid representation enhances feature expressiveness by jointly modeling smooth long-term dependencies and short-term irregular dynamics. The input signals are first denoised using the Empirical Wavelet Transform (EWT) to isolate informative modes from noise, and the SinWavKAN’s hyperparameters are optimized via an Adaptive Tree-structured Parzen Estimator (ATPE). An ablation study confirmed the synergistic superiority of the sinusoidal-wavelet combination over other basis functions. Compared to state-of-the-art KAN variants (e.g., fastKAN, PyKAN, WavKAN), the proposed architecture, named in short Opt-EWT-SinWavKAN, demonstrated a reduction in mean absolute error by approximately 85%–93% across forecasting horizons from 1 to 30 days. The results conclusively show that the Opt-EWT-SinWavKAN provides a more reliable approach for hydroelectric reservoir volume forecasting.
- Patch-based transformer with mixture of experts and conformal inference for curtailment time series forecastingPublication . Heidrich, Mikhail Zimmer; Reis, Mari Aurora Favero; Yamaguchi, Cristina Keiko; Seman, Laio Oriel; Stefenon, Stefano Frizzo; Stefenon, Stefano Frizzo; IEEEThe growth of variable renewable energy sources has increased the occurrence of curtailment in power systems, particularly in regions affected by transmission constraints and limited operational flexibility. Reliable forecasting of curtailment time series, together with well-calibrated uncertainty estimates, is therefore essential to support operational planning, market decisions, and risk management. This paper proposes a curtailment forecasting framework that combines a high-capacity neural time series model with distribution-free conformal prediction to jointly address accuracy and uncertainty quantification. The forecasting backbone is a Transformer-based encoder-decoder architecture enhanced with reversible instance normalization to mitigate distribution shift, patch-based tokenization to reduce sequence length and improve temporal representation, and a Mixture of Experts feed-forward module to enable conditional computation and improved generalization. The proposed method is evaluated against a wide range of state-of-the-art models, including NBEATS, NBEATSx, NHITS, PatchTST, LSTM, GRU, TFT, Informer, and Autoformer, across multiple forecasting horizons. The model achieves an MSE of 0.0059 for a horizon equal to 3h, corresponding to a 13.2% reduction relative to the second-best methods. For a horizon equal to 5h, the MSE is reduced to 0.0119, representing a 29.6% improvement, while for a horizon equal to 10, the model attains an MSE of 0.0461, outperforming NHITS by 22.9%. Ablation analysis confirms the importance of patch embedding, the Mixture of Experts mechanism, and reversible instance normalization. The conformal prediction module achieves an empirical coverage of 91.27% for a nominal 90% level, with adaptive and practically useful interval widths.
- Sparse mixture of experts enhanced transformer architecture for short-term hydroelectric reservoir volume predictionPublication . Seman, Laio Oriel; Yow, Kin-Choong; Stefenon, Stefano FrizzoIn hydroelectric-based systems, effective energy generation planning relies heavily on precise forecasting of reservoir water levels. This paper proposes a novel hybrid forecasting framework that integrates multiple preprocessing strategies with a sparse Mixture of Experts enhanced Transformer architecture for short-term reservoir volume prediction. When evaluated on 19 interconnected reservoirs across two major river basins in southern Brazil using real operational data from the Brazilian National System Operator, the proposed model achieves a mean squared error of 0.062 and a mean absolute error of 0.145. Comprehensive benchmarking against 18 state-of-the-art deep learning methods demonstrates that the proposed approach significantly outperforms existing methods while maintaining computational efficiency through sparse expert routing. Our results confirm that combining diverse preprocessing strategies with conditional computation mechanisms provides superior forecasting accuracy for reservoir management in hydroelectric power systems.