Accessing demand response baselines
- Go to Curtailment > Program Setup > Baseline Configuration.
- Select Baselines to expand the baseline node.
The following are standard baselines:
This algorithm evaluates the last 7 weekdays and discards the highest 2 and lowest 2 usage days on the basis of total energy consumption (total daily kWh). The selection excludes weekends, holidays, curtailment days and test event days. The algorithm averages the remaining 3 usage days, interval by interval, to create a baseline load profile. If the event schedules for a weekend day, the algorithm selects 3 of the last Saturdays or Sundays with respect to the event day. The algorithm treats holidays as weekdays. The algorithm excludes days with missing data and looks for candidate days for up to 7 days in the past.
AvgOfTop5Days (Average of Top 5 Days)This algorithm evaluates the last 30 days and selects the highest 5 usage days on the basis of total energy consumption (total daily kWh). The algorithm averages the 5 usage days, interval by interval, to create a baseline load profile. Holiday and weekend event days follow the same rules. The algorithm does not exclude days with missing data and looks for candidate days for up to 30 days in the past.
This algorithm evaluates the last 10 weekdays and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) between the hours of noon to 6 p.m. The selection excludes weekends, holidays, curtailment days and test event days. The algorithm averages the 3 usage days, interval by interval, to create a baseline load profile. If the event schedules for a weekend day or holiday, the algorithm selects the 3 highest usage days of the last 10 weekend days or holidays. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past.
This algorithm evaluates the last 10 weekdays and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) between the hours of noon to 8 p.m. The selection excludes weekends, holidays, curtailment days and test event days. The algorithm averages the 3 usage days, interval by interval, to create a baseline load profile. If the event schedules for a weekend day or holiday, the algorithm selects the 3 highest usage days of the last 10 weekend days or holidays. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past.
- Highest3of30 (Highest 3 of 30 Days)
The Highest3of30 like-day algorithm selects the 3 days with the highest usage between 2:00 p.m. and 6:00 p.m. within the last 30 calendar days. For weekday event days, the algorithm excludes weekend days, holidays, prior program event days, days with invalid or estimated data, and yesterday. Weekend event days and holiday event days use the same rules except they exclude weekdays from the selection instead of weekend days. There must be at least 3 days available for the service point to participate in the event.
This algorithm evaluates and selects the last 10 weekdays. The selection excludes weekends, holidays and curtailment days. The algorithm averages the 10 days, interval by interval, to create a baseline load profile. If the event schedules for a weekend day or holiday, the algorithm evaluates and selects the last 2 weekend days or holidays. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past, except for a holiday event day in which it looks back for only 2 days.
- Last10DaysV2 (Last 10 Days Version 2)
This algorithm evaluates and selects the last 10 weekdays. The selection excludes weekends, holidays and curtailment days. The algorithm averages the 10 days, interval by interval, to create a baseline load profile. If the event schedules for a holiday, the algorithm evaluates and selects the last 10 holidays. If the event schedules for a weekend, the algorithm evaluates the last 10 weekend days and selects 2 weekend days. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past, except for holiday event days in which it looks back for only 10 days.
- Like-DayBaseline (Like Day Baseline)
This algorithm evaluates and selects the last 8 weekdays or weekend days that are the same day of the week of the selected day. The selection excludes holidays and curtailment days that are of the same day of the week. The algorithm averages the 8 days, interval by interval, to create a baseline load profile. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past.
- PTRResidentialCRL (Peak Time Rebate Residential)
This algorithm evaluates the last 5 weekdays and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) between the hours of 11 a.m. to 6 p.m. The selection excludes weekends, holidays and program event days. The algorithm averages the 3 usage days, interval by interval, to create a baseline load profile. If the event schedules for a weekend or holiday, the algorithm evaluates the last 5 weekend days or holidays and selects the highest 3 usage days. The algorithm does not exclude days with missing data and looks for candidate days for up to 21 days in the past, except for weekdays in which it looks back for only 14 days.
- PTRSmallCommercialCRL (Peak Time Rebate Small Commercial)
This algorithm evaluates the last 10 weekdays and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) between the hours of 11 a.m. to 6 p.m. The selection excludes weekends, holidays and program event days. The algorithm averages the 3 usage days, interval by interval, to create a baseline load profile. If the event schedules for a weekend or holiday, the algorithm evaluates the last 3 weekend days or holidays and selects the highest 1 usage day. The algorithm does not exclude days with missing data and looks for candidate days for up 28 days in the past, except for weekend days and holidays in which it looks back for only 21 days.
This algorithm evaluates and selects the last 1 weekday. The selection excludes weekends, holidays, curtailment days and test event days. The algorithm selects the 1 day to create a baseline load profile. If the event schedules for a Saturday or Sunday, the algorithm selects the previous Saturday or Sunday, respectively. The algorithm treats holidays as weekdays. The algorithm excludes days with missing data the looks for candidate days for up to 7 days in the past.
The following are custom baselines:
- Base Interruptible Program (BIP), Option B Summer
This algorithm evaluates the last 10 days and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) between the hours of noon to 6 p.m. during the summer period, May 1 through October 31. The selection excludes utility holidays, curtailment days and days with rotating outages. The algorithm averages the 3 days, interval by interval, to create a baseline load profile. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past.
- Base Interruptible Program (BIP), Option B Winter
This algorithm evaluates the last 10 days and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) between the hours of 8:30 a.m. to 9:30 p.m. during the winter period, November 1 through April 30. The selection excludes utility holidays, curtailment days and days with rotating outages. The algorithm averages the 3 days, interval by interval, to create a baseline load profile. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past.
- Cafeteria Style Menu (CSM), 1 to 7
This algorithm evaluates the last 10 days and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) between the hours of 1 p.m. to 7 p.m. The selection excludes utility holidays, curtailment days or days with rotating outages. The algorithm averages the 3 days, interval by interval, to create a baseline load profile. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past.
- Cafeteria Style Menu (CSM), 24 Hour
This algorithm evaluates the last 10 days and selects the highest 3 usage days on the basis of total energy consumption (total daily kWh) for a 24 hour period. The selection excludes utility holidays, curtailment days and days with rotating outages. The algorithm averages the 3 days, interval by interval, to create a baseline load profile. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past.
- Last 2 to 5 Weeks Optional Binding Mandatory Curtailment (OBMC)
This algorithm evaluates and selects the last 10 weekdays or holidays. The selection excludes weekends, curtailment days and test event days. The algorithm averages the 10 days, interval by interval, to create a baseline load profile. If the event schedules for a weekend day, the algorithm evaluates the past 10 weekend days and selects 2 weekend days. The algorithm does not exclude days with missing data and looks for candidate days for up to 365 days in the past, except for holidays in which case it looks back for 10 holidays.
This algorithm is unique from all other baselines. It is similar to the User Defined Baseline algorithm in that it is initiated with the creation of an event rather than by a daily task. It is different from all others in the way that baseline days are selected.
The day selection method for this algorithm is iterative, it considers the event period for determination of the peak hourly load over the last 30 calendar days. The usage value equivalent to the peak load over the event period (the seed value) is compared to the usage occurring during the event period two days prior to the event day, eliminating defined holidays, weekend days, prior event days, and any day for which the event period usage is less than 25 percent of the current baseline usage value. The usage from the first baseline day selected replaces the seed value, becoming the current baseline usage value and the comparison process begins again with the next prior day. The event-period usage for every successive day selected is averaged into the baseline usage until 10 days have been selected. The top five usage days among those 10 are selected as the final baseline days. The algorithm averages the five days, interval by interval, to create a baseline.
Weekend day events follow the same procedure, selecting the two highest usage weekend days from the last three. Holidays and event days are not eliminated from selection.
The baseline calculated for program participants identified as generators uses the same methodology for day selection, but selects the lowest usage five days from the last 10 instead of the highest usage days. There is no elimination of low usage days for this type of participant.