Energy Use at Institutional Buildings: Disaggregate Data and Data-Management Issues

摘要

The Federal Institutional Conservation Program includes collection of energy use and energy-related data from individual institutional buildings. Data were obtained from ten states (Massachusetts, New Hampshire, Vermont, New Jersey, Florida, Minnesota, Wisconsin, Texas, Kansas, and Oregon) on almost 15 thousand schools, hospitals, local-government buildings, and public-care institutions. The data were carefully examined and then combined into a single data set. Computer programs were written to summarize the data in various ways (i.e., to prepare tables, cross-tabulations, summary statistics, histograms, and plots) and to identify logical inconsistencies and outliers that might be errors. Both engineering and statistical outlier detection tests were developed and implemented. Then, a clean data set was created that includes only those data elements that passed the error tests. The resultant data provide a wealth of information on energy use and related factors at institutional buildings. For example, mean annual energy use was 155 kBtu/ft exp 2 (1.76 GJ/m exp 2 ) at schools, 370 kBtu/ft exp 2 (4.20 GJ/m exp 2 ) at hospitals, 202 kBtu/ft exp 2 (2.29 GJ/m exp 2 ) at local-government buildings, and 276 kBtu/ft exp 2 (3.14 GJ/m exp 2 ) at public-care institutions. Because data are available (and computerized) for each individual building, it is easy to explore a variety of relationships among such factors as energy use, age of building, fuel prices, primary heating fuel, heating and cooling degree days, hours of operation per year, and the number of building occupants. (ERA citation 06:023715)