Good talking with you this morning.  By all means, talk to Grant Masson about
who else is using the Henwood model within Enron.

Attached are the workbooks I mentioned.  The "Details of Jan and July.xls"
workbook contains the resulting listing from the query I gave you yesterday
and you can see how the supply curve was created from that.  The supply curve
becomes nonsense at points for reasons I believe are related to reliability
commitment constrants, instead of pure economic dispatch, and to the aggregate
reporting problem I described in my note yesterday.

The workbook "Supply Curve.xls" has the simplistic, average supply curve I
mentioned, constructed from fuel and VOM costs.  Depending on the question you
are trying to answer, it may be an approach to consider.

The Henwood Contacts I had in mind are:

Tao Guo, PhD, Senior "Algorithmist"  (916-569-0985) * the one I was thinking
of
Wenxiong Huang, PhD Senior Project Consultant (916-569-0985)
Ajit Kulkarni, PhD, Software Product Manager (916-569-0985) * more a trainer,
but sharp
Cosimo Coscia, Senior Consultant (South Australia) 618-8357-1244 * very
resourceful
Wade Schauer, Staff Consultant, (916-569-0985) * best for questions about EMSS
per se

all have emails, of course.  Template:  tguo@hesinet.com

Also, if you can not get satisfaction, contact Eric Toolson, VP
(916-569-0985).  He has a laconic style, but is very focused on customer
satisfaction and retention.  And he has the pull to make things happen.

Regards,
Michael






>>> Karolina Potter/LON/ECT@ENRON 08/24/00 07:08AM >>> 
Michael, 

I am an analyst in Paul Mead's Continental Power Trading group in London. I am
currently working on the project, which requires the use of EMSS, and
experience some difficulties interpreting the output results. Steven Leppard
from our research group gave me your name as an expert in this system and
consequently the person to contact in case of problems. 

I have been running simulations for the Dutch market and was asked to provide
the traders with some front-end screen graphs in order to interpret the
numerical results. One of the graphs is to show an hourly generation stack and
system's marginal cost, as we only run cost based scenarios. To sort each
station's hourly generation I need its marginal cost. To my knowledge though,
marginal cost is only generated for a systems marginal unit (TransArea
Marginal Units query, Marg_cost unit). Therefore I was sorting the stations
according to the cost which I calculated based on the outputs from Station
Detail by Hour query. The calculation was as follows: 

For each hour, for each generating station: 

"marginal cost" [o/MWh] = (Generation_Cost [o000] * 1000)/Generation [MWh] -
VOM_cost [o/MWh] 

This I thought would include fuel cost and start up costs. However, a marginal
station which I get on the stack as a result of the above calculation is not a
station given in Marginal Station field in TransArea Marginal Units query. I
have also looked into TransArea_Data_Hr table and TransArea_Data table but non
of the costs there match my results. 

Do you happen to know what formula is used to determine Marg_cost and which
outputs I should be using to obtain the right results? 

It might be easier if we could discuss this issue on the phone. In this case
could you please send me your direct telephone number. I am struggling
understanding what is going on and would appreciate your help very much. 

Regards 

Karolina