Estimation of energy intensity by end-use for commercial buildings

Abstract

We describe the results of a study to estimate the energy intensities of major end-uses in commercial buildings in the PG&E service area. These energy intensities are central to commercial sector electricity and natural gas consumption forecasting models that utilize an end-use approach. The use of more accurate energy intensities will result in improved estimates of new power generation needs. To our knowledge, this is one of only three studies using the conditional demand approach for the commercial sector. This technique is less costly than end-use metering a large number of commercial buildings to arrive at energy intensity estimates. We obtained these energy intensity estimates for offices, food stores, restaurants, retail buildings, hotels and motels, hospitals, schools, warehouses, and miscellaneous buildings. We applied the conditional demand technique to monthly energy use, building characteristics, and weather data to produce annual estimates of natural gas and electricity use per square foot of floor space. The consumption estimates are for the following seven end-uses: space heating, cooling, lighting, water heating, cooking, refrigeration, and miscellaneous use. We have compared the results of our study to those of two other conditional demand studies and to other studies where data are available. Presently, there are few metered data available for comparison with our estimates. For most business types and end-uses, the energy intensity estimates appear reasonable. For example, for lighting, the mean energy intensity for each business type is within 20% of all the individual values from all comparable studies. Additionally, for each enduse, the highest energy intensity values occur for those business types that would be expected to show greatest utilization. However, for some end-uses in some business types, there are large discrepancies in energy intensity estimates obtained from various studies. For example, for restaurants, the energy intensity estimates for electric water heating and cooking each vary by more than a factor of seven from minimum to maximum. Additional research is needed to determine if such large variations in estimated energy intensities are real or if they are due to such factors as differences in business types (e.g. liquor stores may be included in retail or food stores), different distributions within a business type (e.g. different ratios of refrigerated to non-refrigerated warehouses), and different floor area definitions.