Direct Digital Control of Small- and Medium-Size Buildings

bustlingdivisionΗλεκτρονική - Συσκευές

15 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

92 εμφανίσεις

Ser
TK1
B92
DIRECT
DIGITAL
CONTROL
OF
SMALL-
AEJn
MEDIUM-SIZE
BUILDINGS
b
=k
*+
A,H,
Efmahdy
and
D.G.
Beattie
INTRODUCTION
The
control
of
heating,
ventilating
and
air-conditioning
(HVAC)
systems
is
changlng
as
a result of
applying
direct
di g i t a l
control
(DDC)
techniques
to
HVAC
control,
mls
report
out l i nes
the
main
features
of
DDC
compared
with
conventfonal
pneumatic
control
and
shows
that,
f or
small-
t e
medium-size
buildings,
the
DDC
system
can
pay
for
i t sel f
within
two
years,
after
which
it
effects
net
savings
over
pneumatic
systems.
This
report
is
a
summary of
a
contract
report
prepared
for
the
Natlonal
Research
Council
af
Canada
by
Systemhouse
~ i d t e d [
11.
CONPARLSDN
BETWEEN
PNEWTLC
CONTROL
AND
DDC
Direct
digital
control
of
HVAC systems
is
the
dfrect
monitoring of
every
system input
(temperature,
flow,
pressure) and
direct
control of
every
system
output
(position,
onlaff)
from
a
central
controller
which
is
a
si ngl e
couputer
or
combination
of
computers.
DDC
is
a
simple
concept,
but
i t s
significance
is
not
obvtous
until
it is
compared
with
traditional
forms
of
HVAC
control.
Traditionally,
the
control of
HVAC
systems
was
based
on
independent
pnewmatic
controllers,
which used
compressed
air
t o
operate
the
dampers
and
valve actuators
t o
control
space
coaditfans
such
as
temperature,
humidity
and
fresh-air
ci r cul at i on.
One
bullding
would
have
several
such
systems,
which
were
controlled
independently.
For
example,
an
air-handling
system
composed
of
two
fans,
three
dampers
and
three
valves
(Figure
1)
would be
cont rol l ed
by
local
pneumatic
controllers
which operated
as
independent
units. Each
cont r ol l er
had
a
s i mpl e
task:
to
maintain
a
constant
set
poi nt
(for
example,
supply
air
temperature)
by
monitoring
and
controllfng
a
very
l i mi t e d
number
of
variables
connected
to
i t
by
means
of
compressed
air
l i nes
whose
pressures
represented t he
values
of
t he
variables.
The
cont r ol
was
adjusted
mechanically
by a
technician
in
the
field,
and,
as
calibratfon
af
the
pneumatic
components
was
rarely
carried
out,
these
systems
often
did
not
control
the
building
effictently.
Because
t he
pneumatic
controllers
were
purely
electromechanical
devices,
their
sophistlcatfon
and accuracy
of control
were
extremely
limited.
A
later
variant
(of
pneumatic
control)
also
employed
pneumatic
central s,
but
w i t h
the
addl t i on
of
a
coqmtes
system.
This
computes
system
monitored
some
additional
points
(for
example,
space
Division
of
Building
Research,
National
Research
Council
of
Canada,
Ot t a wa
%+
Energy
Management
System,
Systemhouse
Ltd.,
Ottawa.
temperatures)
and
either
calculated
new
set poi nt s f or
each
pneumatic
controller
or
allowed
an
operat or
at
a
computer
terminal
to
transmit
manual
setpoints
to
the
pneumatic
controllers.
Although
this
newer
variant
aided
the
bui l di ng
manager
by
providing
more
information
about
butl di ng
conditions
and
performance,
o v e r a l l
effective
control
of
the
bui l di ng
was
st i l l
compromised
by
the
local
pneumatic
controllers.
Each
controlled
polnt
was
still
operated
by
a
pneumatic cont r ol l er
wfth
very
l i mi t ed
sophistication
and
virtually
no
f l e xi bi l i t y.
These limitations
became
more
important
as
ways
to
manage
energy
became
more sophisticated,
Some
WAC
system,
such as
variable
a i r
volume
(VAV)
systems,
required
an
accuracy
of
control
not
attainable in
most
cases
by
pneumatic
controllers.
As
a resul t,
bui l di ng
energy
managers
were
frustrated
by their
knabilitg
to
improve
the
control
strategies
wlthout
rebuilding
the pneumatic
control system
for
each
change.
DDC
has solved
both
problems;. Instead
of
independent
local
pneumatic
controllers,
DDC
uses
control
or
monitoring
points,
each
connected
t o
a
computer
(or
interconnected
computers)
which
reads
the
value
of
each
input
and
transmits
commands
to
each
output
(Figure
2).
The
control
strategies are
implemented by
computer
programs,
which
can
be
changed by
the operator
at
will.
Also,
each
strategy
has
available
to it
t he
value
of every
system
knput
instead of
a
very
l i mi t e d l ocal
set.
In
short,
under
the
DDC
concept,
the
entire
bui l di ng
operates
as
one
integrated
system
rather
than
as
independent
srrrall
systems.
Four
main
results
accrue:
1)
cont rol
can
be
as
simple
or
sophisticated
as desired,
and
can
be
changed
easi l y;
2)
the
system
is
more
reliable
because
fewer
electromechanical
components are
needed;
3)
control
is
more
accurate
because
of
the inherent
greater
accuracy
of
DDC
electronic
components;
and
4 )
energy
is
saved
because
an
over-
all
strategy
elidnates
energy
waste resulting
from
simltaneous
heat i ng
and
cooling,
which
usually
occurs
in
pneumatic systems.
The
abi l i t y
of
DDC
to
accommodate
virtually
any control
strategy
has
had
a
dramatic
impact
on
mechanical
design.
Some
new
mechanical
systems
can
operate in
many
different
modes, depending
on
external
conditions,
space temperatures,
season,
condition
of
storage
tanks,
and
utilitylprfcing
structures.
DDC
allows
such
systems
to
be
operated
continuously
in
t hei r
optimum
modes,
a
standard
which
simply
cannot
be
at t ai ned
by
or di nar y
pneumatic
systems or
even
pneumatic
systems
with
cornput
er
monitoring.
Consequently
,
mechanical
designers
are
now free
to
de s i g n
t he
best
energy
system
for
a
particular
building
with
the
assurance
that
whatever
cont r ol
strategies
they
speci f y
can
be
carried
out.
Each
loop
at
the
remote
processors
can
activate
Itself
independent
of
the
others;
however,
the
most
efficient
use
of
energy
is
achieved
by
controlling
all
the
loops through
the
central
processor.
Scheduling
air-conditioning
and
heating
l oads
and
selectively
dropping
electrical
l oads
if
the total
bui l di ng
power
approaches the
demand
limir
are
t wo
common energy
opt i mi zat i on
features
available.
Other
features,
such
as
optimal
stop/stast,
whtch
calculates
the
ovtimrn
starting and
stopping
times
of
heating/cooling
units
to
prepare
spaces
f or
occupancy
without
wasting
enesw,
are
al so
used
as
part
of
an
over-all
strategy.
Most
of
these
apt i dzat i on
routines
do
not
require
any
additional
hardware
since
they
are
implemented
by
simply
adding
programs
that
sense
existfng
inputs
and
change
the
strategy
for
controllZng
exi sti ng
output
actuators.
The
building
owner
or
manager
who
uses
DDC
effectively
needs
feedback
to
evaluate
h i s
strategies
for
o pt i dz i ng
building
performance.
DDC
simplifies
this process
because
it
contrnually
monitors
each
input
directly
and
has storage
capacfty
to
keep
files
of
the
hi stori cal
data
thus
obtained.
These historical
data
can
be
pl ot t ed
in
colour
on
a
TV
screen
or
summarized
and
printed
i n
report
format
for
management
review. The
most
advanced
DDC
systems
(Figure
3 )
include
a
generalized
report
generator which
can produce
nee
types
of
reports
at
any
time
rat her
than
limit
the
user
to
t he
reports
endsaged
when
the
system
was
procured*
This
feature
of
DDC
i s
particularly
important
since
the
owner's
power
to
change
hi s
energy
strategy
generally
creates
a
need
for
new
repests
on
energy-sensitive
areas
identified
by
continued
use
of
t he
system.
An
ancillary
benefi t
i s
the
ability
of
the
DDG
system
to
include
f aci l i t i es
other
than
WAC.
With
little
increase
in cost,
factors
such
as
control
of
security and l i ght i ng
can
be
added
ta
t he
system,
thereby
enabling
greater
energy savings
and
eliminating
the
need
t o
purchase
separate
systems
for
badge
reading
and
door-lock
control.
There
is
no
doubt
that
DDC
offers
more
effective
energy
management
than
conventional
controls
but,
until
very
recently,
i t s
application
to
HVAC
installations has been
limited
to
large
building
complexes.
Many
small-
and
medi ursi ze
building
installations
do
not use
DDC
mainly
because
of
its
high
ccast[~].
In
the
fol l owi ng
sections a
typi cal
small
bui l di ng
is
analyzed
and
DDC
is
compared
with
pneumatic
control
on
a
cost
and
payback
basf
s.
SMALL-BUILDING
SYSTEMS
The
cast
of
an
WAC
controls Installation is
generally related
to
the
number
of
"points"
t o
be
monitored
or
controlled,
where each
point
is
defined
as
an
analog
o r
digital
i nput
( e.g.,
temperature
sensor,
fan
st at us
switch)
or
analog
or
di g i t a l
output
( e.g,,
damper
position
or
pump
onloff
control.
Each
building
system,
such
as
air
handling,
domestic
hot
water,
or
c hi l l e d
water,
includes
a
certain
number of
points.
A
recent
study
[I]
which
included
det at l ed
anal ysi s
of
a
series
of
bui l di ng HVAC
system,
shorwed
that
a
small-
to
medium-size
bui l di ng
of
about
37,175
m;!
(400,000
sq.
ft
.)
would contain
about
180
poi nt s,
of
which
35%
would
be
analog
inputs,
19%
analog
outputs,
25%
di gi t al
inputs
and
21%
di g i t a l
out put s.
Although
different
building
conf i gmat i ons
and
mechanical
designs
would
affect
t he
distribution
of
poi nt
types,
t he
total
number
of
points
for
a
building
of
this
size
would
usually
be
close
to
180.
DESIGNING
A
DDC
SYSTEM
Given
the
building
layout
and
the
number
of
points
fn
WAC
equipment,
the single
greatest
design
trade-off
is
that
between
centralization
and
distrfbutloa
of
computer
power.
At:
the
fully-
cent r al i zed
extrem,
a
si ngl e
central
computer
cont rol s
a l l
functions
directly
and
all
points
are
wired
to
it.
At
the
other
extreme
( f ul l y-
di s t ri but ed),
a
smaller
central
computer
is
connected
t o a
myriad
of
other
small
computers,
each
of
which
is
wired
to
10
to
20
nearby
poi nt s,
In
t hi s
second
i nst ance
the
central
machine
presides
aver
the
whole
system
and
controls
t h e
points
through
t he
intermediary
of
the
remote
processors.
Each
remte
processor
can
control
a
s i ngl e
WAC
syst em
( e
.g.
,
air-handling
unit,
chiller)
independently.
A
median
approach
is
to
employ
a
moderate
number
of
remote
units
each
of
which
is
wfred
to
50
t o
120
points.
Although all
these
approaches
utilize the
benefits
of
DDC,
the
three
l evel s
of
centralization/distribution
involve
three
factors
that
musr
be
weighed
against
one
another.
The
fi rst
factor
is
the
cost:
of
computer
hardware.
The
ful l y-central i zed
approach
employs
a
si ngl e
processor,
which
is
the
least
expensive
since it
combines
all
the
computing
power
in
one
place
w i t h one
encl osure
and
no
duplication
of
functions.
The
fully-distributed
approach
requfres
the
heaviest
capf t al
cost
for
computer
hardware.
The
second
factor is
electrical
installation
cost.
The
f ul l y-
di st ri but ed
arrangement
yi el ds
the
lowest
installation
cost
because
each
remote
processor
can
be
located
very
close
to
its
points
and
thus
wiring
runs
are
short.
The
fully-centralized
arrangement
may
be
quite
expensive
unless
all
points
are
in
one
mechanical
room.
The
median
arrangement
( Fi gur e
4 ) may
be
the most
economtcal
over-al l because
f our
remote
processors
can
be
used,
one
in
a
penthouse,
one
in
some
other
logical
l ocat i on
such
as a
basement
mechanical
room,
and others on
various
floors of
the
bui l di ng.
The
third
factor is
reliability*
The
fully-centralized
scheme
is
most sensi ti ve to
failure
since
fai l ure
of
t he
single computer
causes
t he
ent i r e
system
to
fail.
Although
the
system
can
be
made
to
fail
safely
,
a
system
failure
is
inconvenient.
The
fully-dis
t r i but ed
scheme
is
least
sensitive
si nce
any
component
computer
can
fail
while
still
leaving
all
the
others
running,
but,
as
previously
mentioned,
the
cost
of
the
computing
equipment
is
highest.
A
median approach
for s ma l l
buildings
makes
good
sense,
A
compromise
on
all
factors
is
established
by
desi gni ng
a
system
consisring
of
a
cent ral
computer
and
four
remote
uni t s.
COST
ANALYSIS:
DDC
VERSUS
PNEUMATIC
CONTROL
The
i ns t a l l e d
cast
of
DDC
systems
has
traditionally
been
hi gher
than
f o r
pneumatic
sys
tens,
especially in
small
installations,
where
the
cost
of
the
DDC
control
processor
is
spread
over
fewer
points.
The
cost
of
a
pneumatic system
tends
to
rise
linearly
with
the
number
of
poi nt s,
as
a large
system
requires
more
independent
local
controllers,
whereas
with
DDC
a
central
processor is
required
even
f o r
system
with
very
f e w
points.
However,
the
rapi dl y
f a l l i ng
cost
of
computing
hardware
has
eroded
the
historical
prfce
difference
between
DDC
and
pneumatic
installations.
For
a
specific
building
[ I ]
of
37,175
n?
(400,000
s q.
ft.),
the
installed
cost
of
a
pneumatic
system
is
about
75%
of
the
cost
of
a
DDC
system.
Although
the
initial
cost
of
a
DDC
system is
higher
than
f or
a
pneumatic
system,
it
can
be
recovered in
a
surprisingly
shor t
t i me.
It
is
realistic
to
assume
that
a
DDC
system
will
yield
a
10X
energ?T
sawing
over
and
above
conventional
pneumatic
control,
due
si mpl y
to
its
more
accurate and
sophisticated
contral,
and t o
its
ability to
provi de
t he
b~i i l di ng
owner
wi t h
information
about
building
performance
and
areas
where
energy
should
be
better
controlled.
Features
such
as
load
shed
and
flexible
schedul i ng
alone
will
produce
large
energy
savings,
and
these
savi ngs
will
increase
as the
owner
becomes more
fami l i ar
~ 5 t h
the
operation
of
the
building.
If
we
assume yearly
maintenance
costs
of
$12,000
and
$10,000
f o r
the
DDC
and
pneumatic
systems
respectively,
and
an
energy
usage
of
322
equivalent
kb7h/dyr.
(30
kWh/sq.ft./yr.)
at
$0.0275
per
kwh
f or
both
systems,
it
will
take
1.4
years
more
for
the
DDC
to
pay
f o r
i tsel f
than
it
will
for
the
pneumatic
system when
used
in t he
but l di ng
under
consideration.
After
that
t i m e
the
DDC
system
will save
money
compared
with
the
pneumatic
controls.
Another
simple
cal cul at i on
shows
t ha t
for
a
three-year
payback
the
DDC
energy
saving
need
be
only
5.J%,
an easi l y
attainable
figure.
CONCLUSIONS
Direct
di gi t al cont ral
is
now
cost
competitive w i t h
pneumatic
control
for
WAC
controls
in
small-
to medium-size
bui l di ngs.
Given
the
other
advantages
of
DDC,
particularly
i t s
abi l i t y
t o
accommodate
changes
t o
cont r al
strategy
and
to provide
det ai l ed
reports of
bui l di ng
performance,
DDC
should
become
the
dominant
technology
for bui l di ngs
in
t hi s
s i z e
range.
Additional
development
is
needed
to
lower
t he
cast
further.
I n
the
exarnple
presented
in
this
paper,
instrumentation
is
the hi ghest
s i ngl e
cost,
largely
because
all
s peci f i ed instrumentation
is
i n d u s t r i a l
grade.
The
development
of
commercial-grade
sensors
and
actuators,
particularly all-electronic types,
which
would
obvi at e
the
need
f o r
an
instrument
air
suppl y,
will
conrsibute
greatly
to the
acceptance
of
DDC.
A
second
area
of
development
lies
in
devi s i ng
computer
program
for
t he
central
and
remote
computers.
As
energy
managers
demand
more
sophistication and
t he
l abour
cost
for
custom
development
ri ses,
comprehensive
and
f l exi bl e software
packages
will
domi nate
the
DDC/HVAC
market.
REFERENCES
1.
Feasibility
Study
of
Small
Building
Direct
Digital
Control.
Prepared
by
Sytemhouse
Lt d.,
for
National
Research
Council
of
Canada
Contract
No.
1SX80-00031,
Uec.
1980.
Note:
I nqui r i e s
concerning
the
avai l abi l i t y
o f
Contract
Report
1SX80-00031
should
be
sent
to
the
Canadian
Institute
f or
Scientific
and
Technical
Informtion,
Building
M-55,
Montreal
Rd., Ottawa,
KIA
OR6.
2.
Elmahdy,
A,, An
Overview
of
Central
Control and
Ilonitoring
Systems
f or
Large
Bui l di ngs
and
Building
Complexes,
Building
Research
Note
No.
159,
Di vi s i on
of
Building
Research,
Nat i onal
Research
Counci l
of
Canada,
March
1980.
SENSORS
8
ACTUATORS
OPTIONAL
CONSOLE
PROCESSING
PROCESSING
I *
.........
C*
.....
l.
I **
........
***I.*..
-------
7
PROCESSI NS
REDUNDANT
OPTIONAL
ALARM
OPERATOR
COLOUR
Dl
SPLAY
LOGGER
CONSOLE
TIGIJRE
3
GE N E R A L
DDC
SYSTEM
BL OCK
DI AGRAM
CENTRAL COMPUTER
FACl
L l T Y
I
1
--
I
BASEMENT
FERST
FLOOR
SECOND FLOOR
THI RD
FLOOR
-
P R f
MARY
PRl
MARY
PRI MARY
OTHERS
AI R
R E MOT E
P R O C E S S O R
REMOTE
P R OC E S S OR
REMOTE
P R OC E S S OR
REMOTE
P R O C E S S O R
FI GURE
4
PARTLY
DI STRI BUTED
DDC
SYSTEM