EnergySmart Grocer

1

• EnergySmart Grocer

2
BPA background

• The Bonneville Power Administration contracted
  Cascade Energy Engineering, Inc. to evaluate the
  GrocerSmart audit tool prior to use in their
  programs.
• Cascade noted eight areas of possible
  improvements to the GrocerSmart 2.0 program.

3
BPA background and GrocerSmart 3.0

• All of these improvements have been implemented
  minus one.
• In addition, GrocerSmart has undergone a number
  of additional improvements, resulting in the
  release of GrocerSmart version 3.0.

4
Outstanding improvement

• Cascade Comment 1 The model overestimates
  savings if multiple interrelated measures are
  implemented.
• Why?
• Savings are calculated for each measure
  individually, using a common baseline
• How?
• If a measure that changes the refrigeration load
  is implemented with a measure that changes the
  refrigeration system efficiency.
• Or
• If multiple measures are implemented that change
  the refrigeration system efficiency.

5
Outstanding improvement - Impact

• Quick Example (site level savings micro look)
• Measure A system improvements that reduce the
  EER from 8.3 to 4.3 (single to high efficiency
  multiplex w/VFD floating head and suction)
• Measure B load improvements that reduce the
  load from 880 BTUH/lf of case to 118 BTUH/lf
  (low temp open to reach-in)

6
Outstanding improvement - Impact

• Quick Example
• Savings
• Measure A 48
• Measure B 87
• Measure AB 135 (no interactions)
• If measure A and B are done at the same time
• Measure AB 93 (including interactions)
• Note this is an extreme example

7
Outstanding improvement - Impact

• Quick Example (program savings macro look)
• Results from a past program
• Frequency of occurrence
• 7 of the 1,258 stores
• Affected savings
• 6 of 51.5 million kWh
• Impact on savings
• lt2 (using the extreme example)
• lt1 (likely impact)

8
Outstanding improvement

• Recommendation To correct his issue, multiple
  system measures or a system with a load measure
  must be analyzed incrementally in sequence or as
  a group.
• Note some system measures are already analyzed
  together.

9
PECI Solution parameters

• Only installed savings, reported to the
  utilities, will reflect interactive impacts.
• Installed savings will not be adjusted for
  interactions on stores that do not require it.
  GrocerSmart will identify and direct users to
  those stores requiring calculation of interactive
  savings.
• Installed savings, previously calculated, will
  never be overwritten.

10
PECI Solution The packaged approach

• Packaged Approach
• Savings are calculated for each interactive
  measure individually (this is done currently).
• Savings are calculated for a single package of
  all installed interactive measures.
• Individual savings are multiplied by the ratio of
  the package savings to the sum of the individual
  savings, thus adjusting the individual savings
  for interactive impacts.

11
PECI Solution The packaged approach

• Installed savings for each measure would be
  calculated as follows
• (GS savings)ARinstalledARinteract
• where
• Ainstalled (installed count)/(GS count)
• Ainteract (GS package savings)/(sum GS
  individual savings)

12
PECI Solution The packaged approach

• Installed savings for each measure would be
  calculated as follows
• (GS savings)ARinstalledARinteract
• where
• Ainstalled (installed count)/(GS count)
• Ainteract (GS package savings)/(sum GS
  individual savings)

13
Packaged Approach - Advantages

• accurate site savings
• the process can be fully automated no user input
  or interaction would be required
• there would be no changes in the way the users
  interact with GrocerSmart.

14
Packaged Approach - Disadvantages

• Every EEM savings will be adjusted regardless of
  whether the EEM has savings interactive with
  other measures.
• This will be addressed by only including measures
  that are logically impacted by significant
  interaction.
• Ie. lighting measures would be excluded, while
  refrigeration load and efficiency measures would
  be included in the interactive adjustment
  process.
•  Reported measure savings may be quite variable
  across the program depending on the groups of
  measures.
•  All EEMs will take on the load shape of the site
  even if the EEM itself has a flat load shape.

15
Other considered methods

• Stepwise Approach
• Calculating savings sequentially in a prescribed
  order
• Advantage
• accurate site savings
• Disadvantage
• Order in which measures evaluated would be
  arbitrary
• Reported measure savings variable across program
• Substantial amount of human interaction for every
  run

16
Example Comparison

• Packaged Approach
• AR 93/135 69
• Measure A 4869 33
• Measure B 8769 60
• Measure AB 93
• Stepwise Approach
• Measure A1 48
• Measure B2 45
• Measure A1B2 93
• Measure B1 87
• Measure A2 6
• Measure B1A2 93

Savings from the previous example Measure A
48 Measure B 87 Measure AB 135 (no
interactions) Measure AB 93 (inc.
interactions)
17
(No Transcript)