Forward Tuning

Forward Tuning is used to adjust the forecast to agree with recent actual observations on the current day. Forward tuning is only useful when real-time data are available, providing actual data for intervals immediately before the first forecast interval. The Forward Tuning parameters specified in the Model Properties dialog are as follows:

• Tune. When this box is checked, the Forward Tuning option is activated.
• Overlap Hours. This is the number of hours to use in the tuning overlap. Model results will be compared to actual values in the resulting overlap period to compute the tuning factors. With sub hourly data, fractions can be entered to represent partial hours (e.g., 2.5 to represent an overlap period of five 30-minute intervals).
• Horizon Hours. Number of hours to phase out the forward tuning adjustment. For example, a value of 12 indicates to phase the adjustment out over a 12-hour horizon. Fractions of hours are also supported here.
• Tune Exclude Load Control Data. If this option is checked, observations marked as Load Control will be excluded from the forward tuning calculation.
• Tune Exclude Load Event Data. If this option is checked, observations marked as Load Event will be excluded from the forward tuning calculations.