KC-03-02-2 (RP-1050) — Inverse Model Toolkit: Application and Testing

Increasing energy prices have brought renewed interest to energy conservation programs, including interest in the development of inverse or regression models for calculating energy savings from whole-building time series data. In response to this need, ASHRAE has sponsored Research Project RP-1050 Ã¢Â€Â”Inverse Model Toolkit (IMT). ASHRAEÃ¢Â€Â™s Guideline 14 for calculating energy and demand savings from energy conservation retrofits has identified three methods for measuring energy savings, including component isolation, before-after measurements, and calibrated simulation. These methods are intended to be guidelines that will serve as a foundation for the development of reliable and accurate measurement of energy and demand savings from energy conservation retrofits. Guideline 14 describes linear, change-point linear, variable-based degree-day, and multivariable linear regression models as the most used models for calculating before-after savings from energy conservation retrofits. ASHRAEÃ¢Â€Â™s Inverse Modeling Toolkit (IMT) is a FORTRAN 90 application for calculating linear, change-point linear, variable-based degree-day, multilinear, and combined regression models. The development of the IMT was sponsored by ASHRAE research project RP-1050 under the guidance of Technical Committee 4.7, Energy Calculations. This paper describes the use of the IMT and the results of testing to ensure that the IMT was accurately calculating the intended models. A companion paper describes the development of the algorithms contained in the IMT (Kissock et al. 2003).