Home >Articles >ANSI Approves RESNET Standard on Method of Test for the Evaluation of Model Calibration Methods

ANSI Approves RESNET Standard on Method of Test for the Evaluation of Model Calibration Methods

Nov 10, 2016

The American National Standards Institute (ANSI) has approved the ANSI/RESNET Standard 1201-2016 – “Standard Method of Test for the Evaluation of Model Calibration Methods” as an American national consensus standard.

ANSI/RESNET Standard 1201-2016 presents an analytical method of test for evaluating software model calibration techniques. The basic concepts for this method are:

  • A building energy simulation program is used to generate pre- and post-retrofit energy consumption data as a synthetic surrogate for data that could be measured in a real building (typically obtained from utility bills).
  • The same simulation program that generates the surrogate measured consumption data is used in conjunction with a calibration technique to produce calibrated and non-calibrated energy consumption data.
  • The calibration technique is evaluated by comparing the calibrated and non-calibrated consumption data to the surrogate measured consumption data along with other metrics.

The test method is useful in several ways including a) testing a single calibration method to see how well it works under a variety of test conditions, b) testing several calibration methods to determine under what test conditions each is best, c) investigating how much, and what kind, of informational content is needed in the synthetic data to achieve good calibrations with different calibration methods (eg. monthly vs daily vs hourly data and availability of different types of submetered or disaggregated data), d) testing with various amounts and kinds of “noise” in the synthetic data, and e) diagnostic testing.

The test method allows users of the standard to construct their own test cases and specifications. However, to avoid the work of creating new test cases, a set of tests and specifications is available from several studies known as the National Renewable Energy Laboratory (NREL) BESTEST-EX reports (Judkoff et. al. 2010, 2011a, 2011b. Information on how to use the BESTEST-EX test specifications.

The standard was adopted after undergoing the RESNET process for the development of consensus standards including a public review and comment process.

The new standard is the result of several years of deliberations by a working group led by Ron Judkoff of NREL.  The development of this Standard was supported in part with funding from the US Department of Energy Building Technologies Office.

To download the standard go to RESNET Standard on Method of Test for the Evaluation of Model Calibration Methods