Research Article
Monthly Electricity Consumption Forecasting Method Based on X12 and STL Decomposition Model in an Integrated Energy System
Algorithm 2
Procedure of the monthly electricity consumption forecasting method based on X12 and STL decomposition models.
| Input: Monthly GDP time series E; monthly electricity consumption time series Y | | Output: Predicted value of monthly electricity consumption | (1) | function X12 | (2) | Decompose GDP sequence E | | | (3) | function stl (ltsObject, s.window, robust = TRUE) | (4) | while ltsObject on season stabilization points | (5) | s.window ⟵ period | (6) | while ltsObject on season alternation points | (7) | s.window ⟵ 2n + 1, n > 3 | (8) | end while | (9) | Decompose monthly electricity consumption sequence Y | (10) | end function | (11) | Predict the trend components | (12) | function VAR | (13) | Estimate the model with the least squares method | (14) | Calculate the current sample output | (15) | end function | (16) | Predict the seasonal components | (17) | function | (18) | Randomly initialize all connection weights and thresholds in the network within the range of (0, 1) | (19) | repeat | (20) | for all do | (21) | Calculate the current sample output | (22) | end for | (23) | until the stop conditions are achieved | (24) | end function | (25) | Predict the random components | (26) | | (27) | Reconstruct the predicted value of monthly electricity consumption | (28) | |
|