How Demand Limiting Works - Metasys - LIT-12011288 - Software Application - Metasys Building Automation System - 6

DLLR Technical Bulletin

Product
Building Automation Systems > Building Automation Systems > Metasys Building Automation System
Document type
Technical Bulletin
Document number
LIT-12011288
Version
6
Revision date
2013-10-15

Demand Limiting executes the following process once each minute:

  1. DL records the meter input value.

  2. DL projects demand for the demand interval. The demand interval is the duration in minutes over which energy consumption is averaged for billing purposes. If projected demand is over the currently active tariff target, DL calculates the required correction. This correction is the goal for DL load shedding.

  3. If the projected demand is above the currently active target, DL sheds the loads in rotation with the lowest priority first, which is priority 10.

  4. If shedding all available priority 10 loads does not satisfy the load shedding goal, DL selects and sheds all available priority 9 loads until the load shedding goal is reached.

  5. If shedding all available priority 9 loads does not satisfy the load shedding goal, the process continues with priority 8 loads, and so forth, until finally the last loads it can shed are priority 1 loads.

    Note: All loads within a load priority either must be shed or ineligible to be shed before DL sheds a load from the next highest priority. For example, if a load is configured at priority 10, it might be shed many times during a single demand interval before a priority 9 or higher load is shed even once. Also, before a priority 8 load is shed, all priority 9 and priority 10 loads already must be shed (or be ineligible for shedding).
  6. If shedding all available DL loads still does not reach the target savings, the following occur:

    • The Status Indicator in the DLLR Focus view displays Trouble. If in Monitor Only mode, DLLR instead indicates its status as Monitor Only Mode Active.

    • The DLLR Status attribute in the DLLR Focus view reads DL cannot meet target.

    • The DL not Shed attribute under the Advanced Demand Limiting section of the DLLR object Universal Input (UI) indicates the unsheddable value.

    Note: If both DL and LR are configured in the DLLR object, the feature sheds loads to satisfy the LR target first. Then, if more shedding is required, DL loads are shed. Loads shed for LR count toward meeting the DL target.

You can shed loads manually to avoid exceeding the demand limit. Choose loads that are not associated with a DLLR object and whose power consumption is reflected in the meter used by the DLLR object.

When a load is shed, the Load extension starts a timer to make certain that DLLR does not control the equipment longer than the maximum time configured by the user. This timer, configured in the Load extension of the object being controlled, is called Maximum Shed Time. When the Maximum Shed Time expires, a Release command is issued to the output. The output might also be released before the timer expires. This occurs if DLLR determines that the current energy consumption allows for a release or an operator releases the output with a manual command.

When the Demand Limiting algorithm runs once per minute, the resultant value is either positive or negative. A positive value (+) indicates a shed situation, whereas the negative value (-) is a release situation. The release part of the algorithm enables the system to release a load before its Maximum Shed Time has expired.

The release algorithm looks for loads that have been shed longer than the Minimum Off Time and are the closest to the expiration of the Maximum Shed Time. For example, if 20 kW are eligible to be released, but the only loads that have been shed are a load of 2 kW with 65 minutes before expiration and one of 19 kW with 66 minutes before expiration, only the load of 2 kW is released.

Note: If a load is released, but the kW consumption does not go up because the load is already off (for example, per schedule), the algorithm does not look for another load to release until the next minute the algorithm runs.

If the load is on a different engine from the DLLR object, and a break in communication occurs between the two (if, for example, the controller with the DLLR object goes offline), the load is released when its Maximum Shed Time expires, unless it is locked/shed.