MicroGrid, a Portland energy services company, and a community college wanted to forge a energy efficiency performance contract. But MicroGrid was swallowed up in so much monitoring data that it was difficult to explain to the client. So they turned to EZ Sim to aggregate and model the college's buildings and then present the information in an easy to understand graphical format that made sense to the college's management.

"This was the only economically viable method with sufficient precision to support our contract." said Terry Egnor, MicroGrid, a Portland energy services company (ESCO). "The method also supported a Quality Control review or savings verification of the retrofit."

MicroGrid had installed short-term monitoring at the site with the intention of establishing a baseline for a performance contract and then to oversee the energy saving retrofit measures. However, the ESCO was literally overwhelmed with monitoring data, making it difficult to extrapolate the short-term measurements to an annual baseline.

In addition, both MicroGrid and the facilities manager wanted the baseline to be able to account for the college's seasonal scheduling changes and variations in local weather, as well as forecast and backcast monthly energy usage.

The ESCO used EZ Sim to easily and successfully model before and after energy usage, and then MicroGrid made a visual presentation that enabled everyone to actually see how the buildings used energy, where the savings would come from and, given the seasonal variations of a college, when the expected energy savings would occur.

This facility is a community college campus with 12 buildings totaling about 318,000 square feet. The entire campus is served by several gas meters and several electric meters, the meters having been added as the campus grew. Complicating the situation, five of the larger buildings are served by a hot water and chilled water loop. It is not possible to isolate the electric and fuel use of any one building. Instead the EZ Sim program is able to sum energy billings and treat the campus as if it were one large facility.

The entire campus received a lighting retrofit and a tune-up of the chilled water loop. After the installation, MicroGrid, a Portland-based ESCO contractor, installed extensive short term monitoring to confirm equipment operating characteristics. The intent was to forge a baseline for a performance contract and to then provide further retrofit services. However, MicroGrid discovered that it was difficult to extrapolate the short-term measurements to an annual baseline. The contractor was literally overwhelmed with monitoring data and, besides, a performance contract based on these types of measurements could have imposed a high overhead cost.

The ESCO and facility manager wanted a performance contract readily understandable to both parties. To do this, the baseline definition had to be able to account for scheduling changes as well as local weather variation. Operations for the college are highly seasonal and will include scheduling changes from year-to-year.

The aggregated billing data shows an orderly pattern for both the electric and gas data, as shown in Figure 1. As a check, this pattern was compared to both pre- and post-retrofit periods. The results were quite similar to an engineering estimate done at that time.

When modeling this and most other college campuses, it is necessary to estimate the mean occupancy for periods corresponding to holidays and vacations, summer school sessions, as well as registration and indoctrination weeks. These varying occupancy situations are readily identified from the academic calendar and from hourly electric data for one of the large buildings.

At first, we included a rigorous analysis for estimating occupancy down to the individual classroom. In retrospect, this exercise was unnecessary. An approximate estimate of the reduced occupancy derived from the academic calendar served quite well.

This close fit to prior engineering work and the ability of EZ Sim to predict both pre-and post-retrofit consumption convinced both the ESCO and the facilities manager at the college to use a model based on the aggregated billings as the baseline for a pending performance contract.

• Adjust ventilation slightly based on calibration.

• Efficient lighting: use proposed connected load of 1.0 W/sqft from lighting audit.
• Upgrade chiller efficiency to 0.65 kW/ton.