System Analysis Advisory Committee Meeting Notes
October 4, 2002 - 8:30 a.m. - 11:15 a.m. .
NORTHWEST POWER PLANNING COUNCIL OFFICES
PORTLAND, OREGON
Attendance: A list of attendees appears at the end
of this document.
Greetings, Introductions and Review of the Agenda.
The
October 4, 2002 System Analysis Advisory Committee meeting, held at the
Northwest Power Planning Council's offices in Portland, Oregon, was
chaired by Michael Schilmoeller and John Fazio, both of the Council staff.
Schilmoeller welcomed everyone to today's meeting, led a round of
introductions, and then reviewed today's agenda. Schilmoeller noted that
copies of his presentation and of the prototype portfolio model are
available on the Systems Analysis Advisory Committee web page (http://www.nwcouncil.org/energy/powerplan/saac/default.htm)
via the NWPPC website. The agenda appears as slide 2 of the
PowerPoint presentation ?SAAC Background.ppt,? available at the web
site. These notes follow the structure of the agenda.
Background
Schilmoeller reviewed the approach and tools we have used in the past
to prepare power plans, and describe what has changed (see ?Background.ppt,?
slides 1-7.)
Audie Nabors asked about the West Coast energy market, noting that, given
the present FERC thrust making the West Coast even more of an integrated
market, reliability issues are going to expand beyond the borders of a
particular state or region. In other words, he said, reliability is
becoming a more expansive issue, and any modeling exercise needs to
recognize that. Nabors thinks our models do take that into
account. Schilmoeller agreed.
?There
are potentially some major changes to the market, due to regulatory
behavior,? said another participant. ?Absolutely,? Schilmoeller
replied.
Schilmoeller describe the proposed portfolio approach (slides
8-20). Schilmoeller then took break for discussion.
In
response to a question about who will care about or use this analysis,
Schilmoeller said that part of the work product is developing an
understanding of what makes certain measures attractive within certain
portfolios, as well as what an optimal portfolio might look like. The goal
is to put into the hands of individual participants information and tools
they might not already have, Schilmoeller explained.
There is a potentially large disconnect between a centrally-planned
resource analysis and getting individual providers to meet the region's
resource needs, observed another participant. ?You?re basically saying
that if you optimize all of the subsystems, you?re not necessarily
optimizing the whole system,? Schilmoeller said. ?If that
occurs, there is definite value in understanding how and why it occurs.?
Objectives
Schilmoeller proceeded to outline the objectives for the SAAC (slides
21-23).
?What
is the real objective?? a participant asked. ?Certainly the Council
has never put out a plan they thought would optimize the regional energy
system; that's been a pipe dream for 15 years. I haven't heard you say
the Council will use this model to develop an optimal portfolio for the
region,? he said. ?What I'm hearing is that this could be a
nice tool for individual participants to use.? that's right ? that's
life, said Dick Watson. ?We don't need such a tool,? said another
participant. ?Before you spend a lot more time, money, and effort
developing this tool, you?d better find out who, if anyone, is going to
make use of it.?
?One
critical question is, how much excess resource do we need?? another
participant observed. ?This kind of tool, which measures risk,
could be valuable for that type of analysis.? ?that's correct,?
Schilmoeller said. ?Our hope is that this group will help us
quantify historical relationships, such as covariance, among key planning
variables and among sources of future uncertainties. If we do nothing more
than that, I think portfolio analysis can make a valuable contribution.?
?I think the idea of running analyses of what individual players are
likely to do, then assessing the sum of all those parts with respect to
regional resources and load, would also be a valuable contribution,?
another participant noted.
?Does
the ?regional data? used in this model include Montana and Wyoming??
a participant asked. Schilmoeller replied that it did not.
Basic Terms and Concepts
Schilmoeller elected to skip a discussion of the
terms and concepts used in decision and portfolio analysis work.
Instead, he quickly reviewed some of the terms and suggested the
definitions be discussed as they are needed. He requested that
participants be aware of the ambiguity of language and asked them to
carefully describe terms they use if there is potential for
misunderstanding.
The Portfolio Model
Schilmoeller took the group on a tour of the prototype portfolio
model. This model is available on the Systems Analysis Advisory
Committee web page.
?When you?re adding resources, there is a capital component,?
Audrey Perino asked. ?How do you incorporate that?? ?We use real
levelized capital cost,? Schilmoeller replied. ?This helps us to avoid
dealing with end-effect calculations.?
In response to another question, Schilmoeller said the loss-of-load
probability calculation used in this simulation bears little or no
resemblance to conventional hourly loss-of-load calculations. ?It
is simply intended to track events where load is not met,? he said, ?and
such events are penalized fairly heavily in this model.? ?Would
we be able to build to a reserve margin in this model?? another
participant asked. ?I believe so, yes,? Schilmoeller replied.
Using the Portfolio Model and Other Council Tools to Make
Decisions
Schilmoeller went through a study flow diagram than illustrates how the
portfolio analysis would be used in conjunction with the Council's other
models, AURORA and GENESYS.
Milestones for the SAAC Process
Schilmoeller went briefly through the main agenda
topics planned for each of the scheduled SAAC meetings (slides
38-43). He then moved on to the final agenda item for today's
meeting.
Action Items
Schilmoeller requested (slide 45) that participants familiarize
themselves with the action items for the SAAC's October 24
meeting. He also requested they examine the agendas for future
meetings and begin locating sources for the kinds of information we will
be needing. This concluded the formal portion of Schilmoeller's
presentation.
?What
I'm worried about is that our solution set is going to be dominated by
the hydrosystem,? Audrey Perino. ?Given that fact, is a regional
solution that looks at hydro as the dominant player in the resource base
useful to anyone on the region except BPA and possibly the Mid-Columbia
PUDs?? she asked. that's just something everyone should think about
prior to the next meeting, she said.
?What
is the time horizon for this analysis?? another participant asked.
?Is it 20 or 30 years, or just the next five or six years?? The
focus of the plan is the next six years, Schilmoeller replied. ?However,
whenever you?re doing this type of planning, you need to know what the
long-term implications of any short-term decisions you make will be. For
that reason, we will be looking out in the future, moving the ?snapshot?
of analysis out in time.?
?When
you say ?mix,? do you mean each part of something that meets some part
of load growth?? another participant asked. ?I mean minimizing the
cost subject to the risk constraint over that time period, Schilmoeller
replied? ?As it is currently configured, the amount of each
resource or measure is constant over the study time period.? The
thing that's hard to capture is the optionality of waiting a year,
another participant noted. ?that's something we've always had
trouble with, so it will be interesting to see if this model can factor
that in.?
The meeting adjourned at 11:15AM. The next meeting of the System
Analysis Advisory Committee was set for Thursday, October 24, from 8:30
a.m. to noon at the Council's Portland offices.
Meeting summary prepared by Jeff Kuechle, NWPPC contractor.
Reviewed by
_______________________
Michael Schilmoeller, Chair
Attendees
|
home