ASHRAE Journal - October 2022 - 28

COLUMN ENGINEER'S NOTEBOOK
FIGURE 3 TMY 3 (normalized) data vs. reality for a year in Portland, Ore.
What ASOS and NOWData are less adept at helping us
understand is what the benefit of a proposed improvement
might be, especially in terms of saving energy. The
NOWData is simply not granular enough, and the ASOS
data is very specific to the realities of a given day rather
than the average conditions.
For instance, if we were to project energy savings for
a proposed improvement based on the hourly data for
a year that had more extreme events than normal, we
might overstate the benefit. But if we use data for a year
that had fewer extreme events than normal, we might
understate the benefits.
That is where the normalized weather data files can
provide benefit.
Normalized Climate Data vs. Extremes
Normalized climate data endeavors to looks at what
subsystems in real time.
* Develop logic that allows the operators to selectively
adjust system and subsystem setpoints to mitigate the
impact of extreme conditions based on experience and
functional tests of the mitigation strategies. For instance:
* A hospitality facility could write logic that temporarily
shifts cooling capacity priority from the guest
rooms to the meeting rooms during the day, when the
guests are likely in the meeting rooms and then back to
the guest rooms when events have ended.
* A health-care facility could write logic that temporarily
resets leaving air temperatures for systems serving
non-patient care areas like administration and the lobby
in favor of maintaining targeted leaving air conditions
for the surgery, ICU and ER.
The logic sequences could be made accessible via buttons
on the summary tables associated with the relevant
systems. The specific parameters invoked by the logic
can be based on functional tests that are executed during
non-critical hours.
Bottom line, extremes will happen. The real questions:
* How prepared are you?
* How bad might it get?
* How often might it happen?
ASOS and NOWData can help you answer these questions
so you can be proactive vs. reactive when the inevitable
occurs.
28
ASHRAE JOURNAL ashrae.o rg O CTO B E R 2022
happened on average. That seems reasonable if one
does not want to over- or understate the benefit of making
an improvement. For a given (unknown future) year,
if you project using normalized data, then " sometimes
you win, sometimes you lose (and sometimes the blues
just gets a hold on you), " but on average you will be
correct.
Common wisdom says that using normalized data
" tends to eliminate or flatten out the extremes. " But
when I took a closer look, I decided I needed to add the
word " kinda " to that statement, as can be seen from
Figure 3.
I say " kinda " because (to my way of thinking) you
would expect normalized data to look more like the
brown bands in Figure 1 and Figure 2, i.e., the TMY
(normalized) data would ride between the extremes.
However, as can be seen from Figure 3, inset A, sometimes
the TMY data does in fact seem to generally follow
the actual conditions. But other times the TMY data
seems to be the extreme (Figure 3, inset B).
Taking a Closer Look
While there have been numerous climate data normalization
approaches used over the years,§
three seem
the most common for use in spreadsheets and modeling
programs. The bin data plot and export feature found
in many electronic psychrometric charts (including
ASHRAE's) use data from one of these sources, allowing
you to get " a visual " of the climate on the chart or export
§Many are documented in footnotes II, II II, and ‡‡‡. I will touch on some here, but will focus on TMY data, which is very commonly applied.

https://www.ashrae.org/

ASHRAE Journal - October 2022

Table of Contents for the Digital Edition of ASHRAE Journal - October 2022