Table 2 Complete set of features after preprocessing and feature engineering.
Feature name | Description |
|---|---|
energy | Target variable (energy consumption) |
hour | Hour of the day (0–23) |
day_of_week | Day of the week (0–6) |
month | Month of the year (1–12) |
year | Calendar year |
season | Categorical season indicator (Winter, Spring, etc.) |
lag_1 to lag_24 | Energy consumption at previous 1 to 24 hours |
rolling_mean_24h | 24-hour rolling mean of energy |
rolling_std_24h | 24-hour rolling standard deviation |
energy_diff | Difference in energy from previous hour |
cumulative_energy | Cumulative energy consumption |
smoothed_energy | Smoothed series using exponential smoothing |
sin_annual, cos_annual | Sine and cosine seasonal encodings for yearly seasonality |
rolling_skewness | Rolling skewness over time |
rolling_kurtosis | Rolling kurtosis over time |
