Truing Up Your Project
Software-based home performance savings estimates have consistently been overestimated; the Program encourages model true up for every project. In order to present your customers with realistic and obtainable energy savings predictions your energy model must accurately describe the home and be ‘trued up’ to their actual utility bills. Your modeling software that you use has the capacity to integrate these bills into the report as you create the model and will show you if some model parameters need to be modified. Other more simple yet adequate methods can be used also, such as working from copies of your customer’s bills and totaling the annual heating fuel and electrical usage then comparing your modeled usage to these totals. The more heating fuel and electric history you have available the more accurate your true-up will be. The minimum period should be 12 months. The model’s predicted energy usage should be within ± 5% of your customer’s actual usage.
To manually evaluate natural gas or electric heating fuel bills:
- Disaggregate heating btus from baseload btus.
- Add up the 3 lowest months of heating fuel usage (summer/shoulder months). For electrically heated homes, avoid months where there is obvious air conditioning usage.
- Divide this total by 3 (or the number of months used), and multiply this by 12. The result is an approximation of your baseload (cooking, water heating, etc.)
- Subtract this annualized baseload from the 12 month total to determine heating related usage.
What can be done if your customers’ bills differ by more than 5%?
Typically this happens when the model predicts usages higher than your customer’s bills. It is not uncommon for models to over predict savings. In this case, the following steps may help:
- Insulation Levels – Review insulation levels in the walls and ceiling/attic. If you have visually inspected the insulation you can use BPI’s Building Analyst Professional Standard’s Effective R-values for Batt Insulation table (CRM Section 13.1, follow link for BA Standard – page 8) to de-rate the R-value. Be cautious in de-rating to a ‘Poor’, especially if you haven’t visually confirmed this. Being conservative in de-rating insulation R-values will likely result in a model that predicts more accurate energy usage and savings.
- Thermostat Setting – Lower thermostat settings. During the wintertime audit you may have observed that the thermostat was set to 70° This does not mean that there is a ∆T of 70° across all exterior walls. Your energy model averages all surfaces adjacent to outside/unconditioned space to predict energy use/loss. Much of these surfaces are well below the set point of the thermostat.
- Estimated Air Leakage – If you were not able to conduct an air leakage test you will have to estimate the building’s leakage. Conservatively use the following guidelines (ACHnatural). If your model isn’t truing up reduce the estimated air leakage:
- Very Tight: 0.10 ACHn
- Tight: 0.35 ACHn
- Average: 0.60 ACHn
- Leaky: 0.85 ACHn
- Very Leaky: 1.10 ACHn
- Conditioned Space Classification – Spaces like basements with little or no specific HVAC supply emitters should not be modeled as Conditioned. Spaces that are classified as conditioned are typically found to be at the set point of the thermostat.
- Hot Water Usage –
a) Lower the water heater temperature set-point.
b) Reduce number of occupants modeled using hot water.
c) Reduce appliance usage of hot water.
For additional information guidance please review the following webinar:
This recorded webinar provides quick techniques and best practices for ‘truing-up’ your energy models to actual energy consumption history to increase savings prediction accuracy, learn how and why model calibration can save you time building your energy model, and understand how the ANSI/BPI-2400 standard supports model calibration and improved accuracy of energy savings estimates.