Page 39 - Zero Net Energy Case Study Buildings-Volume 3

THE J. CRAIG VENTER INSTITUTE
CASE STUDY NO. 12
Renewable On-Site Energy Supply
The solar photovoltaic array spanning the building’s roof and the courtyard space is 500 kW using 1,488 Sunpower E20/327 panels. In 2015, the array produced 800,000 kWh of energy, which is 15% more than the modeled energy consumption for the building. So, the PV system was right-sized for ZNE performance according to the modeled energy consumption, under the assumptions and estimates made for plug load and occupant behavior.
The number of panels is close to the maximum possible on the restricted site. To render the courtyard as pleasant as possible, the number above that outdoor space is reduced to 50% of what would be possible. The daylight studies of the office space determined the spacing of PV panel rows adjacent to the office wing.
The fire marshal effectively determined the number of PV panels above the courtyard space by requiring 50% of the structure be open to prevent the courtyard from being treated as an atrium. The requirement also eliminated the idea of installing clear plastic panels in the openings be- tween PV panels to provide daylighting and rain protection.
Energy Design Analysis and Energy Performance
Modeling versus Post-Occupancy Measurement
Energy Modeling
Energy modeling was carried out continuously during the design phases using eQuest version 3.64. Many of the input parameters, particularly the laboratory equipment load profiles, had to be estimated or researched since no information was available. The heating and cooling system was unusual in its design, and the interaction with the user-operated natural ventilation, preva- lent in this climate, had to be approximated. The result was a reasonable model for the building, which determined an annual modeled energy use of the building to be 700,000 kWh.
The good news for the project team was that this result proved the feasibility of the ZNE goal. Charts showing the modeled annual and monthly energy use by category of load (heating, cool- ing, lighting and equipment) are shown on the following pages.
Energy Use—Actual Measurement and Comparison to Modeling Results
Both the building and solar PV array are separately metered to collect detailed data on the ac- tual energy use and energy production, both as a facilities management tool and to verify ZNE performance. The public utility, San Diego Gas & Electric® (SDG&E®), in this case provides a net meter for the building, which measures the net electric energy flow; namely, the energy used minus the energy produced by the solar photovoltaic (PV) system. The PV array typically has a separate meter that measures just the energy production by the PV system. Therefore, the building owner can check the accuracy of building’s installed metering system by adding the measured production data to the SDG&E net meter data—the result should equal the building’s total metered data for energy use for that period of measurement.
JCVI performed such a check on its metered data system and discovered that it was in error by approximately 20%—the meters were apparently missing some energy-use items. A study of the system identified some building equipment that was not on a metering circuit, accounting for some of the unrecorded energy use, but there remains a significant unexplained load. As of the publication date of this case study, this “unaccounted load” remains unidentified.
Charts showing the measured annual and monthly energy use for the year 2015 are also shown on the following pages. The total energy use is reported based on the SDG&E net meter data (adding the measured PV production to get total energy used). The difference between this amount and that measured by the building’s incomplete metering system is shown as a separate category, “Unaccounted Loads”.
The measured results can be compared with the modeled results to see how the different cat- egories of energy use differ from that in the model. For example, the energy consumed for heat- ing, cooling and pumps is significantly higher than the model indicates.
See the following sub-section, Post-Occupancy: Observations and Conclusions, for further dis- cussion of these issues.
Zero Net Energy Case Study Buildings, Volume 3
25