Short-term wind speed prediction based on FEEMD-PE-SSA-BP

Area of Science:

• Renewable Energy Systems
• Computational Intelligence
• Power Systems Engineering

Background:

• Wind power generation is rapidly expanding but faces challenges due to inherent volatility and instability.
• Grid integration of wind power can lead to power quality issues and operational instability.
• Accurate wind speed forecasting is essential for effective grid management and ensuring power system stability.

Purpose of the Study:

• To address the non-linearity and unsteadiness in wind speed data for improved prediction accuracy.
• To develop a robust and efficient model for short-term wind speed forecasting.
• To enhance the power system's capacity to integrate and manage wind energy.

Main Methods:

• A hybrid approach combining Fractal Ensemble Empirical Mode Decomposition (FEEMD), Permutation Entropy (PE), Sparrow Search Algorithm (SSA), and an Error Back Propagation (BP) neural network.
• FEEMD decomposes wind speed data into intrinsic mode functions (IMFs) based on frequency.
• PE is used to quantify the complexity of each IMF, enabling component merging to reduce computational load. The SSA-BP model then predicts these merged sub-series.

Main Results:

• The proposed FEEMD-PE-SSA-BP model demonstrates high accuracy in short-term wind speed prediction.
• The method effectively handles the non-linear and unsteady characteristics of wind speed time series.
• The simulation results indicate that the model is both computationally efficient and accurate.

Conclusions:

• The FEEMD-PE-SSA-BP model provides a reliable solution for short-term wind speed forecasting.
• This approach significantly improves the accuracy of wind speed predictions, supporting better grid management.
• The enhanced prediction accuracy contributes to the safe and stable operation of power systems with high wind power penetration.