Improved chimpanzee algorithm based on CEEMDAN combination to optimize ELM short-term wind speed prediction.

Area of Science:

• Renewable Energy Systems
• Computational Intelligence
• Time Series Forecasting

Background:

• Global reliance on fossil fuels causes energy depletion and pollution.
• Transitioning to new energy sources, like wind energy, is a global priority.
• Accurate short-term wind speed prediction is vital for power system stability.

Purpose of the Study:

• To develop a hybrid model for enhanced short-term wind speed prediction.
• To address the limitations of low accuracy in existing prediction methods.
• To improve the reliability of wind energy integration into power grids.

Main Methods:

• Wind speed data decomposed using Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN).
• Phase space reconstruction applied to modal components to enhance time series stability.
• An improved Chimpanzee Optimization Algorithm (CHOA) optimized the Extreme Learning Machine (ELM) for prediction.

Main Results:

• The hybrid model demonstrated reduced wind speed prediction errors.
• Significant improvements in the accuracy of short-term wind speed forecasting were achieved.
• Empirical analysis confirmed the model's effectiveness compared to other methods.

Conclusions:

• The proposed hybrid model effectively enhances short-term wind speed prediction accuracy.
• CEEMDAN and CHOA-optimized ELM offer a robust solution for wind energy forecasting.
• This approach contributes to mitigating the impact of wind power fluctuations on power systems.