"Hybrid Supervised Learning Models for Enhanced Accuracy in Renewable Energy Forecasting: Integrating Traditional and Deep Learning Techniques"

EasyChair Preprint 14451

Abstract

The transition to renewable energy sources is pivotal in mitigating climate change and ensuring sustainable energy production. However, the inherent variability and unpredictability of renewable energy generation, particularly from sources such as solar and wind, pose significant challenges for grid stability and energy management. Accurate forecasting of renewable energy output is essential for optimizing grid operations, enhancing energy storage management, and reducing reliance on fossil fuel-based backup systems. Traditional forecasting methods, while useful, often struggle to capture the complex, nonlinear patterns associated with renewable energy production. Conversely, deep learning models excel in handling such complexity but can be prone to overfitting and require extensive computational resources.

The research aims to contribute to the field of renewable energy forecasting by providing a more accurate and reliable prediction model, facilitating better decision-making in energy management systems. The hybrid approach is expected to outperform existing models in terms of accuracy, computational efficiency, and generalization capability across different datasets. This advancement in forecasting techniques has the potential to significantly enhance the integration of renewable energy into the grid, supporting the global shift towards a more sustainable energy future.

Keyphrases: ARIMA, CNN, Grid Stability., Hybrid Supervised Learning, LSTM, Traditional Statistical Models, deep learning, energy management, ensemble learning, renewable energy forecasting

@booklet{EasyChair:14451,
  author    = {Adeoye Qudus},
  title     = {"Hybrid Supervised Learning Models for Enhanced Accuracy in Renewable Energy Forecasting: Integrating Traditional and Deep Learning Techniques"},
  howpublished = {EasyChair Preprint 14451},
  year      = {EasyChair, 2024}}