The OpenEEmeter fits four types of models to the data, scores the model fits, and chooses the best candidate model. Those four types of energy models are:
Payable savings represent the total energy reduction associated with an energy efficiency project. Payable savings are calculated by first using consumption and weather data from the baseline period to establish what consumption during the reporting period would have been in the absence of an intervention. The actual energy usage during the reporting period is then subtracted from this counterfactual (what consumption would have been) to yield payable savings. Payable savings can be used as the performance metric in pay-for-performance settings and can also provide readily available feedback with respect to the savings yields of energy efficiency projects.
Another way to quantify energy efficiency savings is "normal year savings." This is the expected annual savings over a “normal” weather year. The “normal” weather year is an idealized, hypothetical year which is constructed by drawing from observed weather data. This effectively “normalizes” the data by removing some weather uncertainty, and allows for comparison over a number of years.
Comparison groups allow us to compare a group of buildings that received an intervention against a group of homes that did not, but which share as many characteristics as possible for a fair comparison.
Analytics Portfolios are run for older programs or where we have more than a year of post-project data, we call that a Backcast. In Analytics Portfolios we are calculating savings for a program completed or well underway.
This differs from Tracking Portfolios where we have a model of baseline usage and then project model forward in the time after the project.
If, for example, a project started in January and now it’s March, we’ll project for a Tracking Portfolio, the savings from January through March. For Analytics Portfolios, we take a year of data pre project and a year of data post project and then we build a model of both of those. We then calculate a “Normal Year”, which is a compiled set of weather data considered to by “typical”. Data from 1980 to 2010 is used, and they took all the Januarys and picked the most normal looking January, and did the same for each month. We then project the baseline and reporting usage onto that normal year. What the energy usage would have been during that “normal year” for the pre and post periods. That gives us an estimate of what the savings would have been if the weather had been normal.
A site is a physical address or location. Sometimes, a site may have more than one project at that location. A project is defined as an intervention to try and improve energy efficiency. Each site will often have both a gas and electricity meter, but some may only have one or the other.
CVRMSE is an acronym for Coefficient of the Variation of the Root Mean Square Error. The CVRMSE is used to calibrate models in measured building performance. This is a metric that indicates instability in the observed relationship between variables in the baseline period. It is the coefficient of the variation of the predicted input series relative to the observed input series.
Heating Degree Days (HDD) and Cooling Degree Days (CDD) are a measurement designed to quantify the demand for energy needed to heat or cool a building. Generally, a Heating Degree Day is counted when that day’s average temperature is below 65 degrees Fahrenheit, which is the temperature below which buildings need to be heated. A Cooling Degree Day is generally counted when the average temperature is above 65 degrees Fahrenheit which is when the building would need to be cooled.
Load shape is used in energy planning and refers to the distribution of energy requirements over time.
A normal weather year is the hourly outdoor temperature, averaged over multiple years (usually 15 or more). Examples are TMY3, CZ2010 and other weather authorities.
Recurve can display any Project Metadata that is important to you. Typical metadata used are Contractor Name, Measure, Zip Code, or Program. But we can show any information that is important to you.
CalTRACK specifies a set of empirically tested methods to standardize the way normalized meter-based changes in energy consumption are measured and reported. When CalTRACK is implemented through open source software, these methods can be used to support procurement of energy efficiency, electrification, and other distributed energy resources.
CalTRACK methods are developed in an open and transparent stakeholder process that uses empirical testing to define replicable methods for calculating normalized metered energy consumption using either monthly or interval data from an existing conditions baseline.
Learn more on CalTRACK.org.