Performance Modeling of Building Energy Usage: Real Data for REAL Savings

Abstract

The Bonneville Power Administration (BPA) has been using a free
add-in to common workplace software and two easily available data
inputs to verify savings from energy efficiency projects. Regression-
based energy modeling is a technique that uses observed energy use
and measured temperature to model behavior of an existing building.
These empirical models reflect actual building operations and observed
response to temperatures, unlike theoretical energy models (such
as eQuest or EnergyPlus) that assume design conditions and ideal
operation. Readily obtainable data—energy consumption (such as daily
usage from advanced meters or monthly billing data) and recorded
outdoor temperature from a nearby airport—are used with a free Excel-
based tool called ECAM+ to construct models of baseline and post-
retrofit energy use. These empirical models describe real-world building
energy response to outside air temperature, and can be normalized to
typical local temperatures to determine the energy savings from major
retrofit projects. They can also help spot operational issues during the
project performance period. This modeling technique is especially
useful for hard-to-measure, interactive projects such as controls or retro-
commissioning, but can be applied to any project with savings expected
to be greater than a few percent of whole-building energy use.