Frequency-adaptive decomposition with KAN for medium-to-long-term wind power forecasting

Key Points

• Forecasting accuracy improves with frequency-adaptive decomposition in wind power forecasting.
• The methodology employs KAN to refine predictions over a medium-to-long-term period.
• Assessment involved predictive models analyzing wind power data across various conditions.
• Results support the need for advanced techniques to enhance renewable energy forecasting accuracy.