A5050690136- Energy Load and Power Forecasting
- Stock Market Forecasting Methods
- Neural Networks and Applications
- Image and Signal Denoising Methods
- Electricity Theft Detection Techniques
- Sports Analytics and Performance
- Machine Learning and ELM
- Neural Networks and Reservoir Computing
- Power Line Inspection Robots
- Grey System Theory Applications
- Smart Parking Systems Research
- Water Quality Monitoring Technologies
- Sports Performance and Training
- Forest ecology and management
- Advanced Algorithms and Applications
- Metallurgy and Material Forming
Universidade Federal da Bahia
2025
Universidade Federal do Paraná
2016-2022
Load forecasting impacts directly financial returns and information in electrical systems planning. A promising approach to load is the Echo State Network (ESN), a recurrent neural network for processing of temporal dependencies. The low computational cost powerful performance ESN make it widely used range applications including tasks nonlinear modeling. This paper presents Bayesian optimization algorithm (BOA) hyperparameters with its main contributions helping selection algorithms tuning...
Football is the most practiced sport in world and can be said to unpredictable, i.e., it sometimes presents surprising results, such as a weaker team overcoming stronger one. As an illustration, Brazilian Championship Series A (Brasileirão) has historically been shown one of outstanding examples this unpredictability, presenting large number unexpected outcomes (perhaps given its high competitiveness). This study unraveled attack defense patterns that may help predict match results for 2022...
Time series forecasting plays a key role in many areas of science, finance and engineering, mainly for the estimation trend or seasonality variable under observation, aiming to serve as basis future purchase decisions, choice design parameters maintenance schedule. Artificial Neural Networks (ANNs) have proven be suitable linear nonlinear functions mapping. However, ANNs, implemented its most simplistic form, tend loss overall performance. This work aims obtain prediction model short-term...
Several engineering problems are still not solved or there aren't enough hardware, software theories capable of a viable solution.With technology advances and metaheuristics algorithms development, which search solutions inspired in the way nature deals with their problems, such may have solution found.In this context exists Mean-Variance Mapping Optimization algorithm, originally operates over single but has hybrid variant based swarm intelligence incorporates local multi-parent's crossover...