Field Test Evaluation of Conservation Retrofits of Low-Income, Single-Family Buildings in Wisconsin: Audit Field Test Implementation and Results Page: 33 of 84
This report is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
smaller than the furnace gas consumption, was taken as evidence of good data.
Only houses that had anomalous base loads were rejected on data quality grounds.
The few houses that had too little whole house gas data to identify useful base
load patterns were accepted as good data.
Several houses had isolated base loads that were much higher or lower than
normal. In these cases, of which house R24 is a good example, the anomalous
observations were eliminated. Figure 3.2a shows a plot of all pre-retrofit
observations against ambient temperature. A negative base load is obviously
impossible, but it occurs when the measured furnace gas consumption is greater
than the measured whole house gas consumption. The fact that the observations
with negative base loads (2, 6, 13, and 14) also showed higher-than-normal
furnace gas consumption suggests that the furnace gas consumption data are
erroneous. Figure 3.2b shows the data and regression for house R24 after
eliminating the anomalous observations. Note that the regression (model) fits
the data nearly perfectly in Fig. 3.2b but rather poorly in Fig. 3.2a.
House R24 merits further consideration because it illustrates the
sensitivity of this method to anomalous observations. The R2 of the regression
with the anomalous points is 0.86, a value that is considered quite good for
most purposes. However, the standard error of the NAHC (labeled CSE in
Fig. 3.2) is 121 therms/year (9% of the estimated NAHCO). This means that there
is a 90% probability that the true value of NAHC is between 1131 and 1557
therms/year. An uncertainty this large means it is unlikely that energy savings
smaller than a few hundred therms per year can be discerned. Furthermore, the
regression model does not fit many of the observations well, and the base
temperature (Tb) at 70*F is uncommonly large. All these problems disappear when
the anomalous points are removed. Evidently regressions with R2 below 0.9 may
not be useful when trying to measure small to moderate energy savings.
Here’s what’s next.
This report can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Report.
McCold, L.N. Field Test Evaluation of Conservation Retrofits of Low-Income, Single-Family Buildings in Wisconsin: Audit Field Test Implementation and Results, report, January 1, 1988; United States. (digital.library.unt.edu/ark:/67531/metadc740901/m1/33/: accessed August 19, 2018), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.