Renewable energy forecasting using optimized quantum temporal model based on Ninja optimization algorithm.

Area of Science:

• Renewable Energy Systems
• Artificial Intelligence
• Machine Learning

Background:

• Renewable energy systems require accurate forecasting for efficiency and reliability.
• Deep learning methods offer promising approaches to improve renewable energy forecasting.
• Optimization algorithms are crucial for enhancing the performance of deep learning models.

Purpose of the Study:

• To enhance renewable energy forecasting performance using artificial intelligence.
• To investigate the effectiveness of the Q-Transformer Model (QTM) integrated with the Ninja Optimization Algorithm (NiOA).
• To achieve maximum forecasting accuracy by optimizing feature selection in deep learning models.

Main Methods:

• Data preparation including normalization, scaling, and gap handling.
• Application of the Q-Transformer Model (QTM) integrated with the Ninja Optimization Algorithm (NiOA).
• Utilizing NiOA for critical optimization processes and feature selection in deep learning models.

Main Results:

• NiOA demonstrated superior performance compared to other binary optimization algorithms.
• The QTM model with NiOA optimization achieved a high R² performance of 95.15%.
• An exceptional Root Mean Square Error (RMSE) value of 0.00003 was recorded, indicating high accuracy.

Conclusions:

• The Ninja Optimization Algorithm (NiOA) effectively optimizes feature selection for renewable energy forecasting.
• The integration of NiOA with the QTM model significantly enhances forecasting accuracy.
• This approach establishes a robust method for improving the reliability and efficiency of renewable energy systems.