20101130Falkenberg Sur Reb.pdf

to ..
~'l
".Ulû l)l'W 'lO Pt~ 3: 3 ,LdtillJ 'fi~_-l v
\Ll'"t 'l9)~':;~
BEFORE THE IDAHO PUBLIC UTILITIES CÖ~ÖN'"
IN THE MATTER OF THE )APPLICATION OF ROCKY )
MOUNTAIN POWER FOR APPROVAL )
OF CHANGES TO ITS ELECTRIC )
SERVICE SCHEDULES AND A PRICE )
INCREASE OF $27.7 MILLION OR )
APPROXIMATELY 13.7 PERCENT )
CASE NO. PAC-E-10-07
Surrebuttal Testimony of Randall
J. Falkenberg
SURRBUTTAL TESTIMONY OF RANDALL J. FALKENBERG
ON BEHALF OF
THE PACIFICORP IDAHO INDUSTRIAL CUSTOMERS
December 1, 2010
1 Q.
2 A.
3
4 Q.
5 A.
6
7
8
9
10
11 Q.
12 A.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
PLEASE STATE YOUR NAME AND BUSINESS ADDRESS.
Randall J. Falkenberg, PMB 362, 8343 Roswell Road, Sandy Springs, GA
30350. I am the same witness who fied direct testimony October 14,2010.
WHAT IS THE PURPOSE OF THIS TESTIMONY?
My testimony addresses the rebuttal testimony fied by PacifiCorp witness Dr.
Hui Shu on November 16 and on November 24. I update my Table 1
adjustments and address issues related to the screening adjustment, star up
energy, start up O&M, wind integration, and the heat rate adjustment. Unless
otherwise noted, I find the varous PacifiCorp criticisms of all my other
adjustments unpersuasive.
PLEASE SUMMARIZE YOUR TESTIMONY.
My conclusions are as follows:
1. The GRID unit commitment logic contains a serious error,
acknowledged by the Company. Dr. Shu agrees with my proposal to
entirely replace the faulty GRID logic with manual calculations of
daily screens. For puroses of this case, I accept Dr. Shu's screen
modeling methodology, though I disagree with one of her
assumptions.
2. Dr. Shu's opposition to Adjustment 14 (starup O&M) is inconsistent
with her inclusion of these undocumented and unsupported costs in
calculating the daily screens. Remove the impact of starup O&M
increases Adjustment 1 (Commitment Logic Screens) on Table 1 -
SurebuttL.
3. Dr. Shu opposes Adjustment 2 (starup energy) based on two GRID
rus pedormed using the uncorrected GRID logic. Consequently, her
analysis simply measures the random effect of the GRID logic error on
two different scenarios and not the issue of starup energy.
4. Table I-Surebuttal updates Adjustment 2 (Starup Energy) to reflect
the Company's rebuttal GRID run and proposed screens.
1
Falkenberg, Di - Sur
PacifiCorp Idaho Industral Customers
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20 Q.
21 A.
22
5. Adjustment 2 is quite conservative. Inclusion of the energy and
minimum downtime considerations in GRID, as suggested by Dr. Shu
would support a larger adjustment.
6. I accept Dr. Shu's proposal to remove the Seattle City Light Stateline
contract from Adjustments 4 and 5 (Non-Owned Wind Integration).
However, I continue to support the remainder of these adjustments.
This change is reflected on Table 1 - SurebuttaL.
7. Dr. Shu's criticism of Adjustment 13 (Idaho Power Point to Point)
fails to recognize the purose of a balanced pro-forma adjustment.
The Company seeks to include the costs of the new transmission line
as if it came on line in Janua 2010 nearly a year prior to its actual in-
service date. However, in the case of the no longer needed Idaho
Power PTP contract, the Company would continue to include the costs
of the contract in the test year.
8. Dr. Shu's rebuttal of the Adjustment 10 (Heat Rate Adjustment)
addresses a proposal adopted by regulators in Oregon, not my curent
proposaL. My adjustment addresses only the impact of the heat rate
modeling problem at the maximum derated capacity, which Dr. Shu
acknowledges may be valid.
HAVE YOU UPDATED TABLE 1?
Yes. Below is my new Table 1 reflecting my current position on my various
adjustments:
2
Falkenberg, Di - Sur
PacifiCorp Idaho Industral Customers
Table 1 Surrebuttl
Summary of Recommended Adjustents
i. GRID (Net Variable Power Cost Isses)
PacifiCorp Request NPC
A. GRID Commitment Logic Error and Start Up Cost
1 CGmmitment !.egs Ssren4.
1 Commitment Logic Screens1/
2 Sta Up lieFgy 31
2 Start Up Energy 21
B. Long Term Contract Modling
3 SMUD Contrct Delivery Pattrn
C. OATT Wind Integration Cost
4 Nen Owned Inter Hew Wind
4 Non-Qwned Inter Hour Wind
6 NeR O'NneE IRtra Hei:r 'AiR€!
5 Non-Qwned Intra Hour Wind
D. Outage Modeling and Oter NPC Adjustents
6 Lake Side Outage
7 Colstrip Outage
8 JBFuel Adjustents
9 Naughton Outage
10 Heat Rate Adjustent
E. Transmisson Isses
11 DC Interte Cost
12 Populus to Ben Lomond Line Losss
13 Idaho Power PTP Contract
Subtotal NPC Baseline Adjustents .
Allowed. Final GRID Result*
G. Oter Adjustents
44 Cembined CyGle O&M Adjustent
Total Adjustents
Notes
1/ Company Screen Result accepted but increased to reflect 0 start up O&M
21 Based on original (coal value) method. If Min Down Time/GRID value used
Total
Company
Es 10
Jurisdiction
6.36%
5.51%
SE
SG
1,069,701,315 69,200,000
(5BB,42)(34,912)
(3,642,909)(216,134)
(1,676,474)(9,466)
(1,629,483)(96,677)
(1,566,786)(92,957)
(2,041,963)(121,1ãO
(1,367,359)(81,125)
(4,320,031)(256,307)
(2,892,820)(171,631)
(2,163,834)(128,380)
(1,300,710)(77,171)
(2,460,037)(145,954)
(700,273)(41,547)
(1,831,473)(108,661)
(4,766,400)(282,791)
(1,146,067)(67,996)
(842,386)(49,979)
(26,310,536)(1,561,004)
1,043,390,779 67,638,996
(490,000)(29,072)
(26,310,536)(1,561,004)
(1,94,856)(115,507)
3
Falkenberg, Di - Sur
PacifiCorp Idaho Industral Customers
1 Screening Adjustment and Startup O&M
2 Q.WHY is THE GRID SCRENING ADJUSTMENT NECESSARY?
3 A.The GRID model contains a serious logic error that prevents it from correctly
4 determining the most economic star and stop sequence for cycling resources.
5 The problem is so serious that the Company has agreed with the proposal I
6 made in my direct testimony to abandon the GRID logic entirely, to replace it
7 with a manual calculation to determine the optimal daily schedule for cycling
8 resources.
9 Q.
10
HAVE YOU EXAMINED THE COMPANY'S PROPOSED
SCREENING METHOD?
11 A.Yes. It appears to produce results that approach those of the screenIng
12 analysis I have developed. For puroses of this case, I accept their proposed
13 methodology. However, time for review was limited so, I would hesitate to
14 accept it care blanche for all futue cases.
15 Q.
16
17 A.
DO YOU HAVE ANY REMAINING CONCERNS REGARDING THIS
ISSUE?
Yes. The screening methodology considers whether the cost of staing up a
18 unit is offset by the power costs it will avoid. Starup costs have two
19 components - stap fuel and staup O&M. In general, higher staup costs
20 reduce the overall effciency of operation and increase NPC because they
21 prevent certain economic star ups from occuring.
22 As my screening method would change the number of stas, I
23 recommended that these increments to staup O&M be reflected in the test
4
Falkenberg, Di - Sur
PacifiCorp Idaho Industrial Customers
1 year. Dr. Shu opposes this adjustment on the basis that such incremental
2 O&M costs were not originally included in the test year. I disagree with her
3 reasoning. Either star up O&M represents a legitimate test year cost or it
4 does not. If they are legitimate, they should be included in the test year. If
5 not, then they should be excluded from the screening calculation.
6 The Company can't have it both ways - they can't increase NPC on
7 the basis of including staup O&M in the screening calculation, while
8 ignoring the impact eliminating some of these stars on overall revenue
9 requirements. It is puzzling to me that in at least one prior case, the Company
10 did seek to include the incremental star up O&M when it produced higher
11 revenue requirements. In this case, the Company opposed my adjustment
12 when it would lower revenue requirements.
13 Q.
14
DO YOU HAVE ANY OTHER CONCERNS REGARDING THE
STARTUP O&M EXPENSES?
15 A.Yes. I have examined this issue for several years. On numerous occasions I
16 have directed discovery questions at this issue. In all that time, the Company
17 has never once provided any documentation supporting the assumed level of
18 the stap O&M figures they rely upon. Given the circumstaces, I am now
19 questioning whether this "cost" really has any basis in fact. Considering that
20 they do not wish to reflect this cost in the test year, I recommend it be
21 eliminated from the determination of the screens. As a result, I have
22 recomputed Dr. Shu's screening adjustment to reflect the more optimal
23 sequence of stars and stops that would accompany the removal of the
5
Falkenberg, Di - Sur
PacifiCorp Idaho Industrial Customers
1 questionable and undocumented starp O&M expense. This adjustment is
2 shown on Table 1. I also remove Adjustment 14, the original stap O&M
3 adjustment as recommended by Dr. Shu.
4 Start Up Energy
5 Q.
6
7
ON PAGE 25, DR. SHU SUGGESTS THAT NPC SHOULD BE
INCREASED BY $4.7 MILLION IF YOUR START UP ENERGY
ADJUSTMENT IS ACCEPTED. DO YOU AGREE?
8 A.No. First, I'm puzled why the Company would not want to include this
9 additional cost if there truly was a sound basis for doing so. However, review
10 of the calculation of the $4.7 milion figure reveals it is totally lacking in
11 merit.
12 Q.PLEASE EXPLAIN.
13 A.Dr. Shu contends that if starp energy is included in GRID, the minimum
14 downtimes for gas plants should be increased. She then claims that doing so
15 would increase NPC by $4.7 milion. However, her calculation of the $4.7
16 milion is based on taking the difference between two GRID rus with and
17 without the longer downtimes. Unfortunately, her GRID studies are
18 meaningless because they relied completely upon the faulty GRID logic which
19 Dr. Shu has now abandoned. Dr. Shu made no attempt to determine the
20 optimal sequence of stas and stops using a proper screening method. As a
21 result, she seems to be taking the position that two wrongs can make a right.
22 The $4.7 milion figure does nothing more than determine which of the two
6
Falkenberg, Di - Sur
PacifiCorp Idaho Industrial Customers
1
2
3 Q.
4
5 A.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
scenarios is impacted the most by the GRID logic error, not what the actu
impact of modeling minimum downtime would be.
DO YOU HAVE ANY OTHER COMMENTS REGARDING THIS
ISSUE?
Yes. Dr. Shu has ignored the fact that my approach, which values the stap
energy at the cost of coal generation, is very conservative. A more detailed
analysis, which taes account of the actual downtimes and value of
replacement energy as determined in GRID, would support a larger rather than
smaller adjustment. This is because in many cases shut down times are
already long enough to accommodate longer downtimes. Furer, in most
cases, the value of the energy offset (even when reserves and other factors are
accounted for) is much higher than the cost of coal energy. A footnote on
Table 1 shows the value of my adjustment based on an hourly analysis of
GRID rus which considers all these factors which relies on the Company's
GRID rus with the proposed screening adjustment.
In the end, I continue to rely on the original coal based analysis of
staup energy updated for the Company's proposed screens. I continue to
support the coal based analysis because it is simpler and as shown above,
conservative.
7
Falkenberg, Di - Sur
PacifiCorp Idaho Industral Customers
1 OATT Wind Integration Issue
2 Q.
3
4
DO YOU AGREE WITH DR. SHU'S PROPOSAL TO REMOVE THE
SEATTLE CITY LIGHT ("SCL") STATELINE CONTRACT FROM
YOUR WIND INTEGRATION ADJUSTMENTS?
5 A.Yes. The revised adjustments are shown on Table 1. However, I do have
6 concerns regarding whether the SCL contract is compensatory. I find it
7 concerning that the Company may be adopting a strategy of subsidizing
8 wholesale wind generators at the expense of retail customers. I recommend
9 the Commission require the Company to justify the prudence of SCL and all
10 similar wind contracts in the next ECAM filing. In this maner, if the contract
11 costs tur out to be unjustified, ratepayers will be relieved on most of the costs
12 associated with it.
13 I continue to support the remainder of these adjustments for the
14 reasons stated in my direct testimony. As I pointed out before, the Company
15 has had more than six years to have obtained approval to include wind
16 integration charges in its transmission rate strctue. Table 2 (source PUC
17 166) shows the Company's IRP wind integration costs since 2004.
IRPYear Wind Integration Cost Reference to IRP Document
2001RP $4.64/MWh in 200 Dollars 2001RP, Appendix J - Renewable Generation Assumptions, pg 150.
2001RP Update $4.64/MWh in 200 Dollars 2001RP, Appendix J - Renewable Generation Assumptions, pg 150.
20071RP $5.10/MWh 20071RP, Appendix J - Wind Resource Methodology, pg 195
20071RP Update $5.10/MWh 20071RP, Appendix J - Wind Resource Methodology, pg 195
Proxy value of $11.45/MWh.
$8 tax C02 cost Scenario: $9.96/ MWh
20081RP $45 tax C02 cost Scenario: $11.85 / MWh 200IRP, Appendix F. Wind IntelZration Cost Update
$8 tax C02 cost Scenario: $9.96/ MWh
20081RP Update $45 tax C02 cost Scenario: $11.85/ MWh 200IRP, Appendix F - Wind Integration Cost Update
2011IRP $9.70MWh 2010 Wind Integration Cost Study (9-1-2010)
Table 2 PacifiCorp IRP Wind Integration Costs
8
Falkenberg, Di - Sur
PacifiCorp Idaho Industral Customers
1 Heat Rate Adjustment
2 Q.
3
DO YOU HAVE ANY COMMENTS CONCERNING DR. SHU'S
TESTIMONY REGARDING THE HEAT RATE ADJUSTMENT?
4 A.Yes. Dr. Shu addresses an issue not in dispute in this case, albeit one similar
5 to my curent proposal. Oddly, she makes almost no specific comments about
6 my actual adjustment other than to concede that at the derated maximum
7 capacity an adjustment to the heat rates may be waranted (Page 31, lines 18-
8 21, and page 34, lines 3-5). This was exactly what my adjustment does -
9 nothing more or less. This was clearly shown in my workpapers.
10 Examination of my workpapers would have also shown that for 40% of the
11 units, the adjustment is zero or positive (implying an increase to the full
12 derated heat rate). There is no basis for suggesting this adjustment is
13 systematically biased.
14 Q.
15
PLEASE EXPLAIN THE DIFFERENCE BETWEEN THE PROPOSAL
YOU ARE MAKING AND THE ONE DR. SHU ADDRESSES.
16 A.Dr. Shu seems to believe that I modified the entire heat rate curve. On page
17 31, she says PUC would "alter thermal unts' heat rate cures...." The
18 remainder of her testimony, included the figues on page 33 are directed at
19 adjustments made to the overall heat rate cure, not the heat rate at derated
20 maximum capacity. For example, on page 33, she discusses that adjusting the
21 heat cure would, in her view, misstate heat rates below the derated maximum
22 capacity. While her contention is arguable at best, it has nothing to do with
23 my proposal in this case and I won't debate here. Likewise, in the additional
9
Falkenberg, Di - Sur
PacifiCorp Idaho Industrial Customers
1 testimony she fied on November, 24, 2010, (page 35, lines 1-4) she addresses
2 heat rates at loading below the derated maximum. Again, this was not a par
3 of my adjustment. Consequently, I continue to support my adjustment as Dr.
4 Shu has not provided any relevant or persuasive arguents against it.
5 Q.
6
7
ON PAGE 31, DR. SHU CRITICIZES ADJUSTMENTS TO THE
MINIMUM CAPACITY OF GENERATORS. IS THIS PART OF
YOUR PROPOSAL?
8 A.No. Again, she is addressing modeling methods I did not apply in this case,
9 for reasons explained in my original direct testimony and to focus solely on
10 the part of this issue which the Company has already conceded has validity.
11 Finally, it is worth noting that the modeling methods Dr. Shu disparages in
12 ths case were in fact adopted by the Oregon Public Utility Commission
13 ("OPUC") in its Final Order in the recently completed Case, UM 1355. Also,
14 the testimony she presents was also presented in that case and found
15 unpersuasive by the OPUC. That case was conducted over a two-year period
16 and examined a wide range of modeling issues including the interplay
17 between heat rate and outage modeling methods.
18 Idaho Power Point to Point Contract
19 Q.
20
WHY DOES DR. SHU OPPOSE YOUR IDAHO POWER POINT TO
POINT CONTRACT ADJUSTMENT?
21 A.Her reasoning escapes me. On page 38, she acknowledges the Idaho contract
22 was set to terminate based on consideration to the completion of the Populus
23 to Terminal line. Consequently, she seems to acknowledge the contract is not
24 needed afer completion of the new line.
10
Falkenberg, Di - Sur
PacifiCorp Idaho Industrial Customers
1 In ths case, the Company seems confused about the purose of a pro-
2 forma adjustment. A pro-forma adjustment is intended to reflect how all
3 system costs would have changed had the new resource been available for the
4 entire test year under normalized conditions. While the Company includes all
5 of the costs of the new line as if it was in place for the entire test year, they
6 don't wish to consider the fact that par of the value of the line is eliminate the
7 need for varous transmission purchases. Oddly, the Company does agree to
8 exclude some of the low cost transmission purchases which are no longer
9 needed, but prefers to retain this one high cost contract. I see no basis for the
10 distinction between the contracts the Company has agreed to exclude and the
11 one it proposes to continue to include. I continue to recommend this
12 adjustment.
13 Q.DOES THIS CONCLUDE YOUR TESTIMONY?
14 A.Yes.
11
Falkenberg, Di - Sur
PacifiCorp Idaho Industrial Customers