Exponential Moving Average for Time-Series Smoothing in Wind Speed Forecasting

The need for sustainable energy sources is of utmost importance, and wind power is becoming increasingly prominent. Ensuring accurate and dependable wind speed predictions is crucial for tackling energy sustainability issues. This experiment involved using Exponential Moving Average (EMA) to reduce temporal irregularities in a wind dataset time series. The categorization method Long Short-Term Memory (LSTM) was also applied. The investigation concluded with the calculation of a statistically significant R2 coefficient, which particularly reached a value of 0.935. This demonstrates how effectively the recommended method predicts wind speed to maximize renewable energy.