1.3 Model Validation
1.3.1 HERS BESTEST
TREAT complies with Home Energy Rating System Building Energy Simulation Test (HERS BESTEST). HERS BESTEST is a verification procedure developed by the National Renewable Energy Laboratory (NREL) to determine the accuracy and effectiveness of the energy load prediction capability of software tools. The validation methodology consists of comparative testing – in which results from software programs are compared to results from other software programs. The comparative approach includes both ‘sensitivity testing’ and ‘intermodal comparisons’. It uses a wide variety of building configurations and characteristics as test cases for the evaluation. The comparative procedure uses results from three widely-used and well-validated, detailed building energy simulation software programs to develop a range of reasonable results for each of the test cases. The reference programs used to generate the test case results are:
The HERS BESTEST Procedure
1) BLAST 3.0, Level 215: Developed by the U.S. Department of Defense for use in analyzing energy efficiency improvements for their buildings.
2) DOE2.1E-W54: At the time of HERS BESTEST Development, DOE2.1E was considered to be the most advanced of the programs sponsored by the U.S. Department of Energy and the technical basis for setting national building energy codes and standards in the United States.
3) SERIRES/SUNCODE 5.7: SERIRES is a public domain program developed by NREL. SUNREL, the calculation engine behind the TREAT software, was developed as an upgrade to SERIRES..
The results from these three reference programs are then statistically analyzed to determine the 90% confidence interval for each set of test case results. These 90% confidence intervals establish the range of acceptable results for each test case. The National Association of State Energy Officers' (NASEO) National Home Energy Rating Technical Guidelines and the Mortgage Industry's National Accreditation Procedures for Home Energy Rating Systems require that home energy rating software tools "pass" each test for each building configuration that the rating system software intends to evaluate. HERS BESTEST procedures describe two ‘Tiers’ of software test cases – Tier 1 and Tier 2.
BESTEST Tier 1 tests consist of exercising the elements of a basic house with typical glazing and insulation. Specific Tier 1 tests are designed to test a program’s ability produce energy consumption and savings results as described below.
BESTEST Tier 2 tests are more focused on testing a software program’s ability to guide passive solar design, and are not addressed in this document.
Note: Software is considered to ‘pass’ a HERS BESTEST Tier if it passes ALL tests included within the Tier.
Descriptions:Hers BESTEST Tier 1 Test Cases
The following Tier 1 test cases are provided by the HERS BESTEST procedure:
- Case L100: The Base Case Building. This is a 1539 sq.ft., single-story, wood-frame, fully-vented crawlspace home with 270 sq.ft. of single-glazed windows (distributed with 90 sq.ft. on the north and south faces and 45 sq.ft. on the east and west faces). The walls are insulated with R-11 insulation and the ceiling and floor are insulated with R-19 insulation. This is the case against which most other cases are compared to determine if the rating tool can accurately determine energy differences due to changes in building configuration.
- Case L110: High Infiltration (1.5 ach). Exactly the same as Case L100 with the exception of the infiltration rate, which is increased from its base case value 0.67 air changes per hour (ach) to a value of 1.5 ach.
- Case L120: Well Insulated Walls and Roof. Exactly the same as Case L100 except that the wall insulation is increased from R-11 to R-23 and the ceiling insulation is increased from R-19 to R-58.
- Case L130: Double-Pane, Low-Emissivity Windows with Wood Frames. Exactly the same as Case L100 except that the single-glazed windows are replaced with high-efficiency windows having an overall U-factor of 0.30 and an overall Solar Heat Gain Coefficient (SHGC) of 0.335.
- Case L140: Zero Window Area. Exactly the same as Case L100 except that the windows are replaced with wood frame walls having R-11 insulation.
- Case L150: South-Oriented Windows. Exactly the same as Case L100 except that the entire 270 sq.ft. of windows is moved to the south face of the home.
- Case L155: South-Oriented Windows with Overhang. Exactly the same as Case L150 except that an opaque overhang is added at the top of the south facing exterior wall. The overhang extends outward 2.5 feet and is positioned 1 foot above the top of the 5-foot high windows.
- Case L160: East- and West-Oriented Windows. Exactly the same as Case L100 except that all the windows are moved to the east and west faces of the building with 50% (135 sq.ft.) on each face.
- Case L170: No Internal Loads. Exactly the same as Case L100 except that the internal gains are reduced from 68,261 Btu/day to zero.
- Case L200: Energy Inefficient. Exactly the same as Case L100 except for the following:
- Infiltration rate is increased from 0.67 ach to 1.5 ach,
- Exterior wall insulation is replaced by an air gap,
- Crawlspace floor insulation is removed, and
- Ceiling insulation is reduced from R-19 to R-11.
- Case L202: Low Exterior Solar Absorptance. Exactly the same as Case L100 except that the solar absorptance of the roof and walls is reduced from 0.6 to 0.2.
- Case L302: Uninsulated Slab-on-Grade. Exactly the same as Case L100 except that the floor system is changed from a fully-vented crawlspace to an uninsulated, concrete slab-on-grade.
- Case L304: Insulated Slab-on-Grade. Exactly the same as Case L302 except that R-5.4 exterior foundation insulation is added around the slab perimeter.
- Case L322: Uninsulated Basement. Exactly the same as Case L100 except that the floor system is changed from a fully-vented crawlspace to an uninsulated conditioned basement with 1-0" of the uninsulated basement wall and the uninsulated floor band joist exposed. This case is not used for cooling energy load results.
- Case L324: Insulated Basement. Exactly the same as Case L322 except that R-11 insulation is added at the inside of the basement walls and the floor band joist. This case is not used for cooling energy load results.
With the exception of Cases L322 and L324, each of the above test cases is simulated in Colorado Springs, CO to evaluate heating energy loads and in Las Vegas, NV to evaluate cooling energy loads.
Heating Load Results
Table 1 below consists of the 90% confidence intervals for the maximum and minimum ranges of allowable heating annual load predictions produced by the three reference programs compared against the heating energy load predictions of TREAT V3.0.27 in Colorado Springs, CO.
All TREAT V3.0.27 and TREAT V3.0.30 heating load results fall within the 90% confidence intervals required by National HERS standards.
Heating Load Plots
Figure 1 below presents the graphic representation of the data contained in Table 1 above.
Figure 1. Heating load results for test cases L100 – L324 using TREAT V3.0.27 / TREAT V3.0.30 in Colorado Springs, CO.
Cooling Load Results
Table 2 below consists of the 90% confidence intervals for the maximum and minimum ranges of allowable cooling load predictions produced by the three reference programs compared against the cooling energy load predictions of TREAT V3.0.27 and TREAT V3.0.30 in Las Vegas, NV.
All TREAT V3.0.27 and TREAT V3.0.30 cooling load results fall within the 90% confidence intervals required by National HERS standards.
Figures 2 gives results from the cooling load tests using Las Vegas, NV as the climate.
Heating Load Differential Results
Table 3 below consists of the 90% confidence intervals for the maximum and minimum ranges of allowable heating load differential predictions produced by the three reference programs compared against the heating energy load differential predictions of TREAT V3.0.27 and TREAT V3.0.30 in Colorado Springs, CO.
All TREAT V3.0.27 and TREAT V3.0.30 heating load differential results fall within the 90% confidence intervals required by National HERS standards.
Figures 3 gives results from the heating load differential (delta) tests using Colorado Springs, CO as the climate.
Figure 3. Heating load differential results for test cases L110 – L324 using TREAT V3.0.27 and TREAT V3.0.30 in Colorado Springs, CO.
Cooling Load Differential Results
Table 4 below consists of the 90% confidence intervals for the maximum and minimum ranges of allowable cooling load differential predictions produced by the three reference programs compared against the cooling energy load differential predictions of TREAT V3.0.27 and TREAT V3.0.30 in Las Vegas, NV.
All TREAT V3.0.27 and TREAT V3.0.30 cooling load differential results fall within the 90% confidence intervals required by National HERS standards.
Figure 4 gives results from the cooling energy load differential (delta) tests using Las Vegas, NV as the climate.
Figure 4. Cooling load differential results for test cases L110 – L202 using TREAT V3.0.27 and TREAT V3.0.30 in Las Vegas, NV.References
1. Judkoff, R. and J. Neymark, 1995. "Home Energy Rating System Building Energy Simulation Test (HERS BESTEST)," Vol. 1 and 2, Report No. NREL/TP-472-7332. National Renewable Energy Laboratory, Golden, Colorado 80401-3393. (This document also available online at http://www.nrel.gov/docs/legosti/fy96/7332a.pdf)
1.3.2 Other Validation
TREAT heating and cooling load for a single family house similar to the one described in L100A test case of HERS BESTEST was compared to the results obtained with Manual J. TREAT heating and cooling loads proved to be slightly more conservative. Please use professional judgment in applying the results when sizing heating and cooling systems.