[RFI] Utility Automation

[Dave Bernstein]

>>>AA6YQ comments below

-----Original Message-----
From: rfi-bounces at contesting.com [mailto:rfi-bounces at contesting.com] On
Behalf Of Ed -K0iL
Sent: Saturday, September 11, 2004 20:41
To: 'Dave Bernstein'; rfi at contesting.com
Subject: RE: [RFI] Utility Automation


Dave,

Just playing devil's advocate here.

>>>QSL


There are many different SCADA-type systems that could be developed and 
used by utilities but all have one common thing that's missing today: 
Network Infrastructure covering all customers.  Sure, some small
percentage 
of customers today now have cable modems, DSL, and the like, but not ALL
of 
them.  Not even most of them.  Most of them have no reliable or easily 
accessed data link into the home.

>>>For those who are, like me, unfamiliar with the acronym SCADA, its an
abbreviation for "supervisory control and data acquisition", and refers
to an industrial measurement and control system consisting of a central
host or master, one or more field data gathering and control units or
remotes, and a collection of standard and/or custom software used to
monitor and control remotely located field data elements. According to
http://members.iinet.net.au/~ianw/primer.html, contemporary SCADA
systems exhibit predominantly open-loop control characteristics and
utilize predominantly long distance communications, although some
elements of closed-loop control and/or short distance communications may
also be present.

>>>Broadband connectivity in the US is not a small percentage. According
to http://www.websiteoptimization.com/bw/0403/, 45% of all US households
have a broadband connection to the internet, and another 30% have a
lower-speed (e.g. dialup) connection. My assumption is that most if not
all broadband connections are low-latency ("always-connected"), which
for power management applications is likely more important than high
bandwidth.


Here are just three of many possible schemes:

1.  DA or Distribution Automation does not require connections to any
homes 
but it does to various control equipment (switches, cap banks, etc.).
This 
completely lacks any infrastructure today EVEN IF EVERY HOME HAD CABLE 
MODEMS OR DSL.  Some utilities have used wireless to do this in remote 
areas like northern MN.  But the coverage area is limited by wireless 
issues and infrastructure costs of wide area radio systems capable of 
handling wireless data.

>>>What exactly is being automated in this scheme? It sounds like
optimization of the power distribution network, which unaffected by
whether homes do or don't have BPL deployed.


2.  Outage Detection is only useful if a large percentage or nearly
every 
home has a link to provide meaningful info to the utility.  If only
10%-30% 
have it, they won't be able to tell from the data if the outage is a lot
of 
down feeds to homes or a major outage caused by one single device or
feed. 
 So they will still need to revert to the old way of handling this
outage. 
 But if they know it's every home that's out on a given circuit, they
can 
now address that outage much faster and more efficiently than before.  A

BPL Network with new meters everywhere (yes, expensive to do all at
once!) 
could provide meaningful outage detection data.  This will take time to 
migrate towards.

>>>Unless someone's PC happens to be running on an uninterruptable power
supply, it will be incapable of informing the power company via BPL (or
any other communications medium) that the power has failed. You can't
presume the absence of a periodic "hearbeat" message from a home's
connected PC means that home has lost power; the PC may be powered down,
or its operating system may have crashed. One alternative is to deploy a
self-contained battery-powered monitoring device. If BPL or some other
means of internet connectivity is employed, then this device must be
capable of running internet protocols, meaning it contains a
microprocessor, non-volatile memory, and software. You say this is only
useful if nearly every home has this monitoring capability; if so, this
sounds rather expensive. How would BPL make this easier or cheaper?


3.  Energy Mgmt could be done on a limited basis whenever a customer
wants 
it installed to get a lower electric rate.  This is mostly controlling 
overall load through cycling customers' A/C & water heaters OFF/ON 
depending on system load.  It could be made available today to the few
who 
have a reliable network connection (DSL or cable), but the typical
person 
who'd be interested in the savings (lower income households) probably do

not have DSL or cable network connections so BPL would reach these
folks.

>>>According to the above URL, 75% of US homes have an internet
connection. I wonder what percentage of US homes have air conditioners
or water heaters that can be remotely activiated and deactivated via a
PC. Replacing thermostats with HomePlug-equipped units is typically not
plug-and-play because many (most?) are not directly connected to a mains
phase. Water heaters, clothers washers/dryers, refrigerators, and other
household appliances will be equally challenging, though an aggressive
industry standardization effort with consumer incentives for replacement
could reduce this over a ~5 year time horizon.

 
 Watch out what the network providers might start doing once this
becomes 
widespread.  This is just speculation, but there might be additional ISP

costs to allow such commerical access to your home if there are
different 
rates for commercial than normal residential access.

>>>Most ISPs provide flat rate data service. Assuming the above problems
were somehow overcome, the incremental bandwidth generated by realtime
energy management would be "in the noise". I have yet to see an ISP
contract that would preclude realtime control or enable the ISP to
charge higher rates for this application. A single webcam generates
orders of magnitude more network traffic than would the control
applications we've been discussing here.


With BPL, utilities will have "their own network" which they have
control 
over.  No sudden additional charges by outside suppliers.  That's a lot 
more attractive in many ways.  If they could've really done this years
ago, 
they would have.

>>>Utilities can obtain the same flat rate services that other
businesses enjoy. The economic return on deploying one's own network
solely for the purpose of realtime control would be disastrous. Energy
management imposes no quality-of-service requirement that cannot be
handled by existing networks; the communication aspect of realtime
control as you describe it could easily and economically be accomodated
by existing broadband connections, which are already available in nearly
half of all US households. Unless there is unique financial symbiosis
between energy management and BPL, one can't make the other more
economically attractive -- to investors or to consumers. 


For some utilities like public utilities, this may be the only reason
they 
seriously take a look at BPL since they may not be able to get into the 
internet aspect due to regulatory restrictions.  Leasing bandwidth to an

ISP might be an option and would be icing on the BPL cake.  To hams it 
doesn't justify polluting 40-50Mhz or more  of spectrum; but to 
non-hams--the other 99.7% of the population, that's another story.  It's

going to get worse before it gets better, if it does.

>>>From what I've learned in discussions here and in background reading,
the biggest bang for the buck in energy management seems to be reducing
peak requirements by spreading the load over time. This requires no
modifications to household heating/cooling systems or appliances -- it
simply requires the ability to track and report energy usage as a
function of time-of-day, and an electricity pricing structure that
incentivizes consumers to shift their non-critical usage from daytime to
nighttime. As I've said before, this doesn't even require internet
connectivity; a microprocessor-based power meter with a telephone
connection would be sufficient. Yes, more could be done by exploiting
always-connected networks to control in-home equipment, but the cost of
modifying existing infrastructure would be very large. In any event,
always-connected networks are already available; they are not on the
critical path. If power companies are seriously interested in this
approach, why have they not been conducting pilots using existing
networks?

    73,

       Dave, AA6YQ