P2-6. Enhancement of the Whole Building Diagnostician > Background

Developed by the DOE's Pacific Northwest National Laboratory (operated by Battelle), with Honeywell, Inc. and the University of Colorado, the WBD is a production-prototype software package with two modules providing automated diagnostics for buildings based on data collected by DDC systems. See Project 2.4 for a detailed description of Outdoor Air Economizer (OAE) module.

The WBE monitors whole-building and major subsystem (end-use) performance by tracking expected and actual daily consumption as a function of time of day, day of week (and schedules that are correlated with time), and weather conditions. Examples of subsystems that could be monitored, using on-line data from the BAS, include total building energy, electric energy, thermal energy, HVAC energy, and chiller energy. The WBE automatically constructs a empirical model derived from actual past building or system performance and then alerts the user when performance is no longer as good as (or, for retrofits and O&M programs, is better than) past performance. It presents its analysis results as an Energy Consumption Index (ECI) for each day. The ECI is the ratio of actual energy consumption to expected energy consumption. The values of expected energy consumption are generated by a model that performs a statistical analysis of a baseline set of historical data collected from the building or systems. Statistical properties of the expected value are compared to the actual value to determine whether the actual measurement is significantly different.

The existing WBE model used time of day, day of the week, outdoor-air dry-bulb temperature, and relative humidity as independent variables. These variables were found to be good predictors for office buildings, but not for buildings with significant process loads, such as sales volume, refrigeration, cooking, and many manufacturing operations. Energy managers for companies with many similar facilities, such as retail outlets, can benefit from using additional variables, as well as the ability to compare the energy use patterns of similar sites. This project was developed to address these issues.

Back to Previous Page