|
|
|
||||||||||||||||
| Step % of Area Peak Demand |
Cost -- 1997 $/MWh |
|
Step 1 -- 0 to 5% |
50 |
|
Step 2 -- 5% to 10% |
100 |
|
Step 3 -- 10% to 15% |
150 |
|
Step 4 -- 15% to 20% |
250 |
|
Step 5 -- 20% to 25% |
500 |
|
Step 6 -- Over 25% |
1000 |
This supply curve is as good an estimate as Council staff was able to make, but there is relatively little real data and experience to back that up. The sensitivity to this assumption was tested assuming all unserved load has a very high price (5000 mills/kWh). Under these assumptions, almost twice as many aMW of new resources are developed (16000 aMW compared to 9000 by 2010 ? Note that these are the total AMW developed to meet load growth, not the amount developed to replace the dams). As a result, the average amount of annual unserved load (averaged over 50 water years) in 2010 drops from 80-90 aMW to a few tenths. In other words, the amount of new resource development and hence the amount of load that analysis says would go unserved is extremely sensitive to something about which we know relatively little, the value of unserved load.
Ken Dragoon presented information (see presentation) on a spreadsheet reliability assessment model he developed and used while at Bonneville. The model uses At Risk running with an Excel spreadsheet to provide a probabilistic assessment of reliability. The model uses a basic load-resource balance sheet (page 3 in the presentation) with probability distributions for stochastic variables such as hydro conditions, forced outages, and variations in loads. The model performs a large number of load-resource balance simulations letting the stochastic variables vary. The output is a distribution of surpluses and deficits, from which various reliability indices can be calculated.
Ken reviewed several possible indices of reliability and the ambiguity associated with some of the reliability standards (e.g., what does the PNCA standard of one day in 20 years really mean?). He emphasized the importance of expressing the results in terms decision-makers can understand. He suggested translating reliability indices into a surplus or deficit (in MW) for one or more standards.
It was noted that the loads and resources are inputs to the model ? there is no consideration of resource expansion ? and no consideration of economics. However, the advantage of such a model is its transparency. it's relatively easy to understand what's going on. It was suggested that the spreadsheet model be used as a check on the more sophisticated analysis. If there is not good agreement it should be relatively easy to isolate the cause.
Ken reiterated the problem of getting good data with which to do reliability analyses, noting that this is an institutional problem of restructuring.
Pete Swartz described the progress in developing ?GenESys? (Generation Evaluation System), the modeling system being developed for this project. The model combines elements of a number of different Council models and will be able to address generation reliability and the associated economics. Good progress has been made. Pete demonstrated the operation of the model. It appears to run quite efficiently and provides a great deal of flexibility in specifying the inputs, run parameters and output information. However a great deal of work remains to be done. Pete is now looking at mid-February for having the modeling relatively complete. A group of those advisory committee members with a particular interest in the modeling will be meeting with Council staff to review progress and assess if this is the proper direction for the modeling.
A question was raised regarding what to use for and how to treat contracts for export. The recommendation was to use what in the PNUCC Northwest Regional Forecast and/or the White Book, acknowledging that these are likely to be increasingly incomplete or inaccurate.
Another question was how to treat those contracts in a reliability analysis. Can we assume that in a period of tight supply in the Northwest, Northwest generators with contracts to supply loads outside the region would be able to ?counter-schedule,? i.e., make purchases in the export region (depending on the availability of capacity) to serve their contract loads while using their NW generation to meet regional needs??
A caution was expressed that that might not be the case. To do so might mean that the out-of-region customer would have to accept service over a transmission path for which the supplier does not have firm rights. The customer might not be willing to accept this.
A question was raised that if you saw a cold snap coming, wouldn't you get on the phone and try to firm up the ability to counter schedule? Yes ? but you might not be able to.
The comment was made that we have to be very careful about assuming that we can max out the intertie capacity when we need it.
It was suggested that we start out with the firm loads and resources in
the NW and then see how much you would have to lean on other sources like
displacing firm exports, making imports and curtailing load.
Wally Gibson presented a rough draft aimed at structuring the policy and institutional issues associated with assuring an adequate and reliable power supply during the transition to a fully competitive power market (attached). It was acknowledged that that transition was underway but how far and how fast it might proceed is an open question.
Most were comfortable that markets, left to their own devices, would develop to address the adequacy/reliability issue. There are concerns, however, that markets won't be left to their own to develop. This could be for reasons unrelated to reliability. Or, it could be that extreme prices and/or supply disruption associated with a period of tight supplies could result in interventions that might be counter-productive and that might prevent markets developing to the point that they can adequately address reliability issues.
One member stated the goal of this study roughly as follows: ?What are the impediments to getting from where we are today (partial restructuring) to a world where there are perfect markets, perfect contracts. And what do we do about it??
Wally is seeking comment from the committee and others on this draft.
The role of quantitative analysis in this issue is open to question. The goal is to be credible and understandable to decision-makers. How sophisticated does the analysis have to be to meet that goal?
The next meeting of the advisory committee was schedule for the afternoon of February 18th, 1:30 ? 5:00 at the Council's offices.
home