20141023Helman Direct.pdf

(mq 343-7s00
(208) 336{912 (Fax)
McDevitt & Miller LLP
Lauryerc
42) West BannockSteet
P.O. Box 2fi4..83701
Boise, Idaho &3702
Octobet 23,2014
rar_nr l r"
201! 0[T 23 PH tr: 52
l'lAijr; 1-,.; .,,. , -
Ul I L I T I r: S . Cr i,i ifi I tj$He[ilF. McDevitt
DeanJ. $oe) Millet
Via llaad Delivery
JeanJewell, Secetary
Idaho Public Utilities Commission
472W. l7ashington St.
Boise,Idaho 83720
Re: Sierra Club/ IPC-E-14-18
Deat Ms.Jewell:
Enclosed fot filing please find the odginal and fline (9) copies of the Testimooy and exhibit of Udi
Helman. One copy of the Testimony has been designated as the "Reporter's Copy." In addition, a
dkk gontaining MS Wod venion of the Testimony and a PDF version of the exhibit is enclosed fot
the Reportet.
If you have any questions, please do not hesitate to contact me.
Kindly returo a samped copy.
DJM/hh
ORIGINAL
Dean J. Miller (ISB No. 1968)
McDEVITT & MILLER LLP
420 West Bannock Steet
P.O. Box 2564-83701
Boise,ID 83702
Tel: [phone]
Fax: [phone]
i o e@mcdevitt-mi11er. com
Matt Vespa
CA Bar #222265 (Pro Hac Vice)
Sierra Club
85 Second St., 2n'l Fl.
San Francisco, CA 94105
[e-mail]
Tel: [phone]
Fax: [phone]
Attorneys for Sierra Club
IN THE MATTER OF IDAIIO POWER
COMPA}IY'S APPLICATION TO
IMPLEMENT SOLAR INTEGRATION
RATES AND CIIARGES.
l-rf 6t- ,t t/" !1i-=.c1"- "-., "'
._1,, )
?c,q OcT 23 PFl h: 52
il,,'.i i;_1
'' .: , '
UT lLlf iIii C{t,,.'i r i,.::i..; ii ; i'
BEFORE TIIE IDAHO PUBLIC UTILITIES COMMISSION
) cAsE NO. TPC-E-14-18
)
)
)
)
)
DIRECT TESTIMONY
OF TJDI HELMAN ON BEHALF'OF SIERRA CLUB
October 23,2014
1 Q. Please state your name, business address, and occupation.
2 A. My name is Udi Helman. My business address is 155 Jackson Street, #1306, San
3 Francisco, CA94111. I am currently an independent consultant.
4 a. Please describe you work experience.
5 A. Previously, I was ernployed by BrightSource Energy (a developer of solar thermal
6 power plants), the Califomia Independent System Operator (CAISO), and the
7 Federal Energy Regulatory Commission (FERC). At the CAISO, I worked on a
B major renewable integration study.l At Brightsource Energy, I spent time
9 comparing solar valuation studies, and also served, and continue to serve, on the
10 Technical Review Committee for an NREL study of comparative solar valuation. I
7t have a PhD in energy economics and systems analysis. My curriculum vitae is
72 appended hereto as Exhibit 401.
13 a. On whose behalf are you testifying in this case?
74 A. I am testifuing on behalf of Sierra Club.
15 a. What is the purpose of your testimony?
t6 A. Sierra Club requested that I evaluate the Idaho Power Company's (IPC) Solar
L7 Power Integration Study (henceforth, the IPC Study). Sierra Club also asked me to
18 comment on acfual experience with solar integration and the calculation of solar
L9 integration costs and the extent to which it may inform solar integration costs in
20 Idaho. While I comment on the IPC Study and review the results from other solar
2t integration studies in western utilities, I have chosen to discuss in some length the
1 California ISO and GE Energy, Integration of Renewable Resources: Operational Requirements and
Generation Fleet Capability at20% RPS, August 31, 2010, htp://wwwcaiso.com/Documents/Integration-
RenewableResources-OperationalRequirementsandGenerationFleetCapabilityAt20PercRPS.pdf.
Helman, Di
2
Siena Club
L
2
3
4
5
6
7
8
9
10
7t
12
13
L4
15
!6
t7
1B
L9
20
2L
22
23
a.
A.
evidence to date in the CAISO market, simply because there is some transparent
data on leading indicators of actual integration costs as solar production has
increased over the past two years. I also refer to a summary report jointly prepared
by the CAISO and NERC on measure to prepare for wind and solar integration.
However, I also stress that each power system operator needs to analyze its own
system.
Please summarize your testimony.
The testimony will discuss aspects of the methodology in the IPC Study, including
potential improvements (since I have reviewed comments by the TRC, I do not
repeat all their concems). tn particular, I note results from integration studies that
use a "net load" calculation of reserves needed for wind and solar integration, and
observe that, while requiring more complicated statistical models, these results are
likely to be lower than the method used by IPC to add the reserves calculated
separately for load, wind and solar. I also note that while the use of production cost
simulation is now a conventional methodology for calculating integration
requirements and costs, the IPC model is not transparent, and hence more difficult
to evaluate by the TRC or other third parties. To provide evidence of how other
power systems are adapting operations to higher levels of wind and solar
integration, I review the results from the CAISO markets. Based on my review, it
appears that the procurement and market prices for frequency regulation have not
increased at all despite the rapid increase in wind and solar production. [n addition,
the procurement and modeled shadow prices for the real-time "flexi-ramp"
constraint have actually decreasedin2}l4 despite the continued rapid increase in
Helman, Di
5
Sierra Club
L
2
3
4
5
6
7
B
I
10
tt
t2
13
74
15
76
17
1B
!9
20
2t
22
23
a.
A.
solar production. These indicators do not suggest that integration costs will not
increase in the future, but they are suggestive that the available flexibility of the
CAISO resource mix and other measures taken to improve forecasting, control,
visualization and optimization of available resources have limited such costs to
date. There may be lessons for IPC in solar integration that do not require joining
an ISO or even an EIM. Finally, I recommend that the current study results are
considered to be still in draft, with IPC required to respond to the prior comments
from the TRC. Also, the role of the TRC should be expanded to assist in the
evaluation of integration solutions. A broader group of stakeholders and experts
could be reached if there is additional transparency in study inputs, as has been
evidenced in some of the processes established to evaluate renewable integration in
California.
What are the operational requirements and costs associated with renewable
integration?
Renewable integration costs are, in principle, any additional costs that a utility or
buyers in a regional power market would incur due to changes in system operations
to accommodate variable energy resources - wind and solar. As evaluated by IPC
for use in determining avoided costs under PURPA, these costs would be net of any
avoided energy and capacity costs due to renewable energy; that is, they would be a
deduction from the PURPA avoided cost rate. Integration costs are controversial
because they are difficult to measure, both on actual power systems and in
simulations of future conditions Interestingly, recently in Califomia, integration
costs have been of interest in the competition among renewable developers,
Helman, Di
4
Sierra Club
1
2
3
4
5
6
7
B
9
10
t7
t2
13
t4
15
16
t7
1B
L9
20
2L
22
particularly to higher cost but less variable renewable technologies that are
interested in improving their comparative valuations in utilityprocurements (such
as geothermal and concentrating solar power with thermal storage). I do not oppose
calculation of integration costs using long-term simulations for planning or
renewable portfolio standard (RPS) procurement purposes so that utilities can more
effectively evaluate the mix of renewable and other technologies for their power
systems, but I also believe that when assigning such costs explicitly to long-term
contracts (as a price deduction) or even through wholesale markets, there should
also be a comprehensive and reasonably transparent effort to find cost-effective
integration solutions. We see this basic approach being implemented in Califomia.
To date, as discussed below, the evidence in Califomia seems to be that integration
costs are not yet significant, probably due to many operational and market
modifications to improve flexibility, even though the power system has added about
4,500 MW of solar and 4,000 MW of wind since 2012, for a total of about 5,000
MW of solar and 6,000 MW of wind as of September 2014.
When power systems reach operational limits, despite measures to improve
operational flexibility, renewable energy tends to be curtailed, which is a cost borne
by the developer or the off-taker depending on the type of project and the terms of
particular contracts. At that point, integration costs could refer also to the lost
renewable energy at contracted rates. The IPC study does not consider those costs,
so I don't comment on them here in much detail.
What are your views on the general framework for the Study?
Helman,
5
Siena Club
a.
Di
3
4
As noted in Phil DeVol's testimony, the Study followed guidelines that are
reflective of the current state of modeling in this area. The general framework of
using a statistical model to calculate additional reserye requirements, which are then
used in a production cost model to develop changes in production costs, has become
fairly standard. There are a number of modeling methods that could be improved
and would likely reduce the calculated integration cost within the models used.
Some of these have already been noted by the TRC and are repeated in other
testimony, so I will only address a few issues, such as "netting" of load, wind and
solar reserve requirements, where I can provide additional information.
It is worth noting that integration modeling continues to develop into more
complex types of analysis that I discuss briefly below. But what has also been
missing from research is validation that simulation results for future years turn out
to be correct, and if not, to adjust the modeling accordingly to improve accuracy. In
the past, this was because the models were evaluating conditions on power system
that did not yet exist; but many power systems are catching up to, if not exceeding,
the levels of wind and solar modeled in earlier integration studies. So there is the
ability now to do some model validation. One can infer this from the empirical
results I present from the CAISO market, which suggest that actual integration costs
to date are less than might have been predicted by earlier integration modeling.
What are your views on how the reserve requirements were calculated for the
model?
Helman,
6
Sierra Club
5
6
7
I
9
10
TL
L2
13
L4
15
L6
L7
18
t9
20 a.
2t
22
23
Di
A.Several TRC members have already criticized the method of "stacking", rather than
netting, reserve requirements due to load, wind and solar.2 That is, the IPC
methodology adds the reserve requirements calculated individually for variability
and forecast error in load, wind and solar - which are then used in the production
cost model to determine the change in production costs - rather than calculating the
reserve requirement around the "net load" operating point, that is, the load minus
the wind and solar in each five-minute interval modeled. In actual practice, there
will be times when the three types of variability could offset each other, reducing
the reserve requirement. This is shown in the CAISO 20% RPS study cited above
in footnote 1, where in several figures -- o.8., Figure A-9 on page A-15 and Figure
A-14 on page A-21-- the reader can see that in some hours the maximum load-
following or Regulation requirements for "load + solar" or o'load * wind" are
slightly lower than for "load" alone, or that "load * wind + solar" is sometimes
lower than "load * wind". While conducting this kind of statistical modeling is
more complicated than calculating requirernents separately for load, wind and solar,
there is no question that the netting calculation should produce lower total reserves.
And in fact, this is how the CAISO and other Balancing Authorities (BAs) would
actually operate their systems - by forecasting the net load operating point for each
dispatch interval. There is a caveat for this example - the CAISO model in
question is different in several other ways from the IPC model.3 But this caveat
2 See, e.g., IPCo Response to Frist Discovery Request of Sierra Club, Request No. 12 (comments of Jobn
Crider, OPUC Staff, Kurt Myers, Idaho National Lab.
3 For example, the CAISO statistical model uses a Monte Carlo simulation to evaluate the interactions among
large numbers of draws from load, wind and solar forecast errors, and then utilizes the maximum requirement
at some threshold, e.g., the 95%opercettile of all draws in that iteration of the modeling.
Helman,
7
Sierra Club
2
3
4
5
6
7
I
9
10
\t
t2
13
t4
15
76
L7
18
L9
20
Di
2
3
4
5
6
7
B
9
10
L7
18
t9
ZO
21
22
23
a.
A.
notwithstanding, it illustrates the general impact of netting in lowering forecasted
reserve requirements.
Is production cost simulation an appropriate tool to use when evaluating
changes in future integration costs?
Yes, production cost simulation has become a primary tool in renewable integration
studies because it allows for examination over long periods (e.9., one or multiple
years) of scenarios with significant changes on the power system, such as high
penetration of wind and solar, while representing detailed operational and
transmission constraints. The production cost models also have limitations. They
can be computationally intensive, thus limiting the number of runs that are feasible,
particularly for larger regional systems. The results are sensitive to input
assumptions, such as forecast fuel costs. As renewable integration issues become
focused on sub-hourly details of power system operations, such as load-following,
frequency regulation, and primary frequency response, the frontier of production
cost simulation has shifted towards the capability to analyze some of these
capabilities internally to the models or by coupling them with other subhourly
models. In addition, there is research into representing forecast errors directly into
the simulations using stochastic methods. To continue a theme raised above, it
will be important to continuously validate these simulation methods to ensure that
they are reflective of acfual operational and market outcomes.
Is the IPC production cost model similar to those used in other studies?
As stated by IPC, the production cost model is an internally developed model and
hence is difficult to compare directly to other models. IPC was helpful in providing
Helman, Di
8
Siena Club
LI
L2
13
t4
15
76
a.
A.
L
2
3
4
5
6
7
I
I
10
LL
t2
13
L4
15
L6
L7
18
L9
20
2L
22
23
24
25
a.
A.
results from the model in response to daa requests, and these results show some of the
operational changes needed for solar integration.
Eow do the solar integration costs in the Study compane to other solar integration
studies?
Given this methodological evaluatiorq the next issue is whether the resulting integration
costs are comparable to those from other studies, and if so, why. As noted in Phil DeVol's
testimony, the solar integration costs found in the study are comparable to those in other
studies that use similar simulation methods. There are a range of methodologies and some
charges are developed more transparently than others, but most appear to fall into a ftrnge
of $1 - $6lNIWh, depending on the quantrty of solar modeled, with higher costs for higher
quantities. Some ofthese integration cost forecasts are used for integrated resource
planning studies, while others are used to adjust avoided cost rates.
A PV integration study performed for NV Enerry calculated integration charges in the
range of $3/MWh forthe first 150 MW of PV to about $7l[\{Wh for 1,042 MW of PV, and
an additional $lilvfwh for PV curtaiLnent costs in the latter case. PacifiCorp in Utah has
proposed a solar integration charge deducted from its avoided cost rate for QF contracts of
$2.18iIvIWh for ftacking solar and $2.834,twh for fixed solar. APS has calculated a solar
integration cost of $2.08/I\dWh for 1,038 MW of solar, and $3.04/MWh for 1,669 MW of
solar. @lack & Veatch, Solar Photovoltaic (PV) Integration Cost Study, conducted for
APS, 2012.) BPA has calculated an integration charge of $0.21lkW-month for 23 MW of
solar. LADWP has calculated an integration of $7.64lMWh for up to 614 MW of solar.
(Cited in Los Angeles Deparhnent of Water and Power, 2013 Power Integrated Resource
Plan. December 16, 2013.) Tri-State has calculated a charge of $2.181N4Wh for 20 MW of
solar. (Tri-State Generation and Transmission Association, Inc. Integrated Resowce
Plon/Electric Resotrce P/ane, November20l0.) TEP has calculated a
Helnan, Di
Corrected Pg.9
Sierra Club
2
3
4
5
6
7
I
9
10
L7
L2
13
L4
15
L6
L7
18
19
20
2L
22
23
a.
A.
$5.20AdWh cost for the first 100 MW of solar PV, with an additional $1.10/l\dwh
for each additional 100 MW. (Tucson Elechic Power. 2014 Integrated Resource
Plan. April 1,2014.)
However, like the IPC study, these are all models attempting to estimate
future integration costs. We don't know whether these estimates are correct or
incorrect for the particular systems modeled until there is more operating
experience wittr wind and solar on these systems.
What power systems can we look to for examples of solar integration at high
penetrations?
There are a number of power systems around the world that have already
experianced high and increasing levels of solar generation, whether utility scale or
distributed. These ftmge from island systems, such as Hawaii, to large US states,
such as Califomia, and, of course, Germany. Of these, in the U.S., only Califonria
also has a transparent wholesale market operated by the California Independent
System Operator (CAISO), which gives more insight into how market prices and
costs are evolving with renewable integration.
How much renewable enerry is now on the California IS0 power system,
measured in the aggregate?
Under the33% RPS, the Califorria load-serving entities are required to achieve
33olo renewable energy, not including hydro, by 2020. Compliance could comp
earlier than2020 due to the potential for changes in financial incentives (e.g., the
investment tax credit), which is leading solar projects to come on-line earlier. Of
these load-serving entities, the Califonria investor-owned utilities are jtrisdictional
Hetnan, Di
Corrected Pg. 10
Sierra CIub
a.
A.
t
2
3
4
5
6
7
B
9
10
7t
L2
73
t4
15
L6
t7
18
L9
20
2t
22
to the Califomia Public Utilities Commission (CPUC), with the municipal utilities
subject to oversight by the California Energy Commission (CEC). The CAISO
market discussed below has historically covered the investor-owned utilities and
some of the non-generation owning municipal utilities, accounting for a little over
80% of California load, but more recently has expanded to include the Energy
Irnbalance Market (EIM).
Of interest here is the expansion in solar and wind production within the
CAISO footprint, and in particular solar production, which began to increase more
rapidly after 2012. In20l2, the maximum actual hourly solar output on the CAISO
grid was about 522MW, while the maximum actual hourly wind plus solar joint
production was 2575 MW. The growth since then has been rapid. Within the
CAISO footprint, in September 2014, the maximum hourly average solar
production, jointlybetween PV and concentrating solar power (CSP), was 4,889
MW, and the total maximum joint solar and wind production was 6,777 MW.
Under the33% RPS mandate, the final capacity of new renewable resources will be
around 20,000 MW, with solar capacity expected to consist of about half of that.
Hence, the CAISO market will continue to face rapid expansion in wind and solar
production, and the findings on renewable integration to date, as discussed below,
may not necessarily be sustained. At the same time, as also discussed below, new
flexible resources may be coming on-line concurrently, making baseline
measurements diffi cult.
What is the experience to date at the California ISO with solar integration?
Helman,
l1
Sierra Club
a.
Di
t
2
3
4
5
6
7
B
9
10
tt
72
13
L4
15
L6
17
1B
t9
20
A.First, I note that I no longer work at the CAISO, and hence my observations here
are based on public data and reports. Like other ISOs and RTOs, these days the
CAISO is in a fairly continuous process of operational and market reforms to
accommodate the increasing level of utility-scale and distributed renewable
generation as well as new technologies that can support wind and solar integration,
such as fast responsive storage technologies. a Recent improvements that have
been publicly shared are a regulation and load-following ramping forecast tool for
the systern operators, which allows them to check the ramping range in each
dispatch interval up to several hours ahead. The CAISO is also evaluating three
different methods for setting Regulation requirements while meeting the new
NERC Balancing Authority ACE Limit (BAAL) control performance standards.
Earlier, the CAISO also moved from a set hourly Regulation requirement to a
variable forecast hourly Regulation requirement more consistent with actual uses of
Regulation Up and Regulation Down over the operating day. As discussed below,
the CAISO has also included a new "flexi-ramp" constraint in its real-time dispatch
algorithms on difflerent time-frames, which allows for additional ramping capacity
to be hold to account for uncertainty about "net load". In addition to these factors,
the ISO control room likely has additional forecasting improvements and software
and visualizationtools which assist renewable integration, but that are not
necessarily public knowledge.
a Some of these new capabilities are documented in "2013 Special Reliability Assessment: Maintaining Bulk
Power System Reliability While Integrating Variable Energy Resources - CAISO Approach." A joint report
produced by: the North American Electric Reliability Corporation and the California Independeut System
Operator Corporation, November 201 3 Helman, Di
t2
Sierra Club
1.
2
3
4
5
6
7
8
9
10
\t
t2
13
!4
15
t6
17
1B
t9
20
2L
a.
A.
Complementing these actions at the CAISO are a range of policies at other
California state agencies and by the utilities to support renewable integration on the
bulk power network, and most recently, to adapt to the new operating conditions at
the distribution level caused by the penetration of distributed resources. These
include adding a category of "flexible capacity''to the CPUC's resource adequacy
program requirements and implementation of the State's storage mandate to load-
serving entities. Since these types ofpolicies were not evaluated in the IPC Study, I
do not review them here, except to note that they will also affect the rate of
penetration and, indirectly, the costs of renewable integration.s
What indicators of wind and solar integration costs can we observe at the
California ISO?
The CAISO does not specifically break out "integration costs" as a category of
market costs, and in fact, this would likely be very difficult to calculate, since some
of these costs are borne through non-market payments, often called "uplift". Hence,
we can observe the changes in the costs of certain ancillary services and any other
operating requirements or constraints that are priced transparently. However, in
each case, there are caveats which are duly noted.
Of these, I would include changes in frequency regulation prices, which are
procured at CAISO through three separate products - Regulatio, Up, Regulation
Down, and Regulation Energy Management-as well as a recent operational
constraint called the "flexi-ramp" constraint, which allows for additional units to be
s I use the tenn "indirectly''because policies such as the California storage mandate have goals other than
immediate operational applications, such as tech:rology promotion/market tansfonnation. Hence, it would
not be accurate to count the costs of meeting the storage mandate as a renewable integration cost.
Helman,
13
Sierra Club
t committed orredispatched to provide a wider ramping range during real-time
2 operations.
3 Q. Has the California ISO procured any additional frequency regulation over the
4 past 3 years? And have prices increased?
5 A. Frequency regulation is an ancillary service procured to balance the power systern
6 through automatic controls on intervals shorter than the dispatch intervals. ln the
7 CAISO, dispatch instructions are sent every 5 minutes, and Regulation is dispatched
8 on a 4-second basis in between those instructions to balance deviations. The
9 CAISO has both a Regulation Up product and a Regulation Down product, allowing
10 for resources to determine which direction to provide automated response.6 These
LI markets procure Regulation capacity (MW) and set a price in terms of $AvIW. In a
12 series of earlier studies, the CAISO forecast that it would procure additional
13 Regulation at levels of wind and solar penetration below what is currently on the
t4 CAISO power system.T However, based on the public data shown in the table
15 below, the CAISO does not seem to have procured any additional Regulation Up or
16 Regulation Down over the past three years even as wind and solar penetration has
t7 increased fairly rapidly. The price of Regulation seems to have increased slightly in
18 2014 compared to 2013, but from a very low base. Moreover, the Regulation
t9 clearing price was higher in previous years before the increased penetration of new
20 wind and solar, reflecting its close correlation to the price of natural gas.
6 Most other ISO markets have a combined Regulation service, where resources provide a regulating range
around a set point.
7 For example, the study conducted in 2010 estimated that Regulation procurement would increase by I I - 38
%, depending on the season, for less additional wind and solar than is currently on the CAISO grid.
However, since that model was deployed, further adjustments have been made to scheduling time-lines and
assumed forecast errors, so these earlier estimates could be lower if the same study was repeated today.Helman, Di
t4
Sierra Club
Market clearing prices Regulation quantities
Reg Down ($/MW)Reg Up ($/M!V)Reg Down (MW)Reg Up (MW)
2012 4.39142 5.644762 351.3854 332.9629
2013 3.254272 4.557149 323.8746 336.5561
2014
(Jan.- Sept.)
3.911s45 5.284641 329.9904 344.s816
a.
A.
L
2
3
4
5
How much "flexi-ramp" has the California ISO procured since implementing
the constraint and how much has it cost?
The CAISO established the flexi-ramp constraint in December 2071, primarily to
support upward ramping requirements in the real-time market. While not initially
established for purposes of variable energy generation, this constraint is commonly
interpreted as an indicator of integration needs, since it could be expected that
increased intra-hourly variability would require additional such reserves (sometimes
called a ramping reserve). The CAISO operators can adjust the constraint to
address operational needs, and the CAISO has also reviewed the performance and
adjusted the minimum and maximum allowable capacity made available to
minimize the effects on the energy dispatch markets. For example, the CAISO
reduced the initial maximum capacity from 900 MW to 600 MW in January 2014,
which has reduced the costs associated with this constraint. This adjustment also
demonstrates the difficulty in discerning what is actually causing integration costs -
external factors, such as increased wind and solar production, or intemal changes to
system operations and optimization algorithms.
Helnan, Di
l5
Sierra Club
6
7
B
9
10
tt
t2
13
L4
15
L5
t7
1 The average quantity and "shadow price" on the flexi-ramp constraint is shown in
2 fhe table below. What is clear is that over 2012,2013 and 2014 to date, the CAISO
3 has procuring varying amounts of flexi-ramp capacity, and that the price has also
4 varied. Most notably, the quantities and costs have decreased coincidentally with
5 increasing solar penetration in 2014.
Average hourly quantity (MW) in each real-time interval forward in time
0 t5 30 45
2012 352.1875 353.7335 353.9592 354.4962
2013 480.4835 480.9865 481.2001 480.6074
2014
(Jan. to Sept.)
39s.1439 395.1988 394.9072 393.8919
Average hourly shadow price on the flexi-ramp constraint ($/IvIW)
0 15 30 45
2012 6.807843 3.065125 1.878161 1.435347
2013 6.610604 3.0st626 2.793484 2.307141
2014
(Jan. to Seot.)
t.509974 1.78778 t.743549 1.694296
6
7
8
9
10
77
72
13
74
15
t6
These results are shown graphically below, but with more details on how monthly
costs have varied. Note that the y-axis on the right hand side is in $millions, not
$nvfw.
Helman, Di
t6
Sierra Club
5,000
4,500
4,000
3,500
3,000
2,500
2,000
1,500
1,000
500
0
$3,00
$2.s0
$2.00
$1.s0
$1.00
$0.s0
$0.00
o*t'+ot' .*f C o,d ."f "t'.."-..,f *S *.Js,-g "****9 oJ
-Solar
Max Production (MW)
-Flexi-ramp
Cost f$millions)
,/\A
\I
I
\A 4-f\
^
t
\-Y\l\t \Lv\/\/\/\
v \./\
\
t
2
3
4
5
6
7
I
I
10
LL
t2
13
t4
a. Has the frequency of negative prices increased?
A. Another indicator of integration cost is overgeneration, which results from "must-
take" generation (renewable and non-renewable) exceeding demand in particular
intervals. Similarly to the question of reserves, there arc aran1e of operational and
market adjustments that can be used to avoid or minimize overgeneration. Negative
prices have been increasing in the CAISO market, and are forecast to increase in
correlation to increasing renewables. However, as noted above, other corrective
measures are also being developed and evaluated, such as improvements in
conventional generator flexibility, increased deployrnent of energy storage, new
types of flexible demand response for ramp mitigation, and adjustments to the future
resource mix in the renewable portfolio, such that we can expect the frequency and.
magnitude of negative prices to fluctuate over the years in response to these changes.
Helman, Di
t7
Sierra Club
L
2
3
4
5
6
7
I
9
10
7L
72
13
t4
15
t6
77
18
t9
20
2L
22
23
a.
A.
What do you conclude about integration costs to date in the California ISO
markets?
As I noted above, there are factors that I cannot control for when answering this
question. However, there is no question that it is difficult to find clear trends in
prices of market products and non-market costs that would be leading indicators of
integration costs. This seems to indicate that the CAISO power system has been
sufficiently flexible to date to integrate wind and solar without incurring significant
integration costs. In other words, increased penetration of renewable resources has
not resulted in increases in actual integration costs as predicted in earlier modeling.
Since this finding is not what was expected a few years ago - for example,
observers might have anticipated, based on the earlier CAISO integration studies, at
least a few dollars of added Regulation costs - it provides a further data point from
a high penetration wind and solar power systern that atreasonably high and growing
penetrations on a large regional power system, solar integration does not appear to
cause additional integration costs of any significance. The lesson for smaller
systems is to take the operational lessons leamed from these larger systems, such as
the range of operational and forecasting tools and market reforms cited in the
CAISO-NERC review referenced above; moreover, theymay also provide
confidence that closer integration with a larger regional market, such as an EIM,
could support lower costs of integration, as evidenced by the apparently low costs
of integration to date in the fairly large CAISO market.
What does this review of the Study and experience elsewhere with solar
integration lead you to conclude?
Helman, Di
18
Sierra Club
a.
L
2
3
4
5
6
7
8
9
10
11
12
13
t4
15
76
L7
1B
t9
20
2t
22
23
A.First, I commend IPC for undertaking a detailed solar integration model that has
provided the foundations for further analysis, and also for being forthcoming in
response to data requests. Based orr the review of the Study and additional material
on the experience in the CAISO (based on public data sources) I have discussed
above, I support doing additional methodological review to address the modeling
issues identified above, along with the modeling issues identified by the TRC, and
possibly more detailed simulation of system operations than done in the current
Study, along with more transparency over the results and more opportunities for the
TRC or other stakeholders to recommend sensitivity analysis to reflect the adoption
of operational improvements. As discussed above, there is simply a very broad and
growing set of operational improvements and solutions that are being observed and
evaluated in power systems around the country, and which will directly affect the
integration of wind and solar resources.
Status of IPC Study. With respect to the current IPC Stud5 my
recommendation is that it is considered a draft document pending further review by
the TRC. My understanding is that solar developers have already, or are currently
negotiating contracts with IPC for Commission approval that include integration
costs consistent with those estimated in the IPC study. Therefore, formal adoption
of the current IPC study is not a necessary prerequisite or constraint on deployment
of additional solar resources in Idaho. Instead, adoption at this time could result in
precluding consideration of additional modeling sensitivities and operational
solutions that could reduce actual integration costs, as observed in regions like
California that are already experiencing high levels of solar penetration.
Helman, Di
t9
Siena Club
L
2
3
4
5
6
7
B
9
10
Value of External Expert Review. Having both responded to, and
participated in expert advisory committees, similar to the TRC for the IPC Study, I
believe that there is a high value to such review. In the next phases of this study,
the TRC should be tasked with reviewing the state-of-the-art in simulation used to
calculate integration costs, but also with balancing increasing complexity of
simulation with the task of finding a reasonable integration cost glven the rapid
changes in system operations and technology. The TRC could also support review
of low-cost operational and software solutions being used elsewhere to support
renewable integration. The current thinking in Califomia and elsewhere is that
operational flexibility is not necessarily in short supply on the power system if the
right signals are provided. This has been demonstrated in several markets, where
new ancillary service products and market pricing rapidly elicited new types of fast
response capabilities from conventional generation (primarily combustion turbines),
demand response and storage, and in some cases led to reductions in the quantity of
ancillary services procured.8
Transparency. Another aspect of my experience at the CAISO and
afterwards is the benefit of transparency about modeling to a broader group of
stakeholders, including renewable project developers who need incentives to
provide the capabilities needed by systern operators. As one example, the CPUC,
the CAISO, and the CEC have developed a stakeholder process for evaluating
renewable integration simulations under the CPUC's long-term procurement
I A notable example is the PJM RegD, or "fast regulation" market. See the description and reyiew of the first
year of market operations here: htp://www.pjm.cono/-/media/documents/ferc/2013-filingsl20l3l0l6-er12-
1204-004.ashx. Note that the number of resources following the RegD signal has continued to increase
following this report.Helman, Di
20
Siena Club
tl
L2
13
1.4
15
t6
t7
1B
t9
20
2L
1 planning proceeding, in which a data-base of the Califomia systern and rest of
2 WECC using public data is posted serni-annually, and is used by many parties,
3 including the national laboratories, to test operational requirements and solutions
4 using production cost models and other models - and to innovate in such modeling.
5 While there are limits to this process (notably that the actual network models used
6 by the CAISO are not being modeled), at the least it has helped to shape discussion
7 and provided a benchmark for analysis. The Califomia utilities also continue to
B develop their own proprietary models, which are used in planning and procurernent.
9 a. Does this conclude your testimony?
10 A. Yes it does.
Helman, Di
2t
Sierra Club
E. Udi Helman
155 Jackson St., Apt. 1306
San Francisco, CA 94lll
Udi@HelmanAnalytics. com
E)(PERIENCE
Independent Consultant, energy and environmental analysis, San Francisco, CA (512013 -
present). Advising companies, trade associations and research institutions on policy, regulatory
and market issues as well as quantitative analysis, primarily related to valuation and integration of
renewable energy and energy storage.
Various current public roles and selected consulting clients:o Chairman, Market Analyics Working Group, Energy Storage Association (ESA),
Washington, DC, November 2013 to present. Organize and run monthly calls as well as many complementary activities. Speakers
to date from ERCOT, PJM, CPUC, CAISO and other organizations.I Initiated ESA Frequency Response Task Force and continue as co-chair. Multiple roles in aillual ESA conference, Washington, DC, June 4-6,2014o Co-char, IEEE Task Force on Storage Modeling, September 2014, ongoing. Technical consultant, EPRI Energy Storage Integration Council, ongoing.o Technical Review Committee, NREL, Valuation of concentrating solar power and other
solar technologies, August 20l l to present.
o Policy Advisory Commiffee, California Energy Commission/DNv-Gl project on modeling
concentrating solar power with thermal storage, August 2012 onwards (project final report
under review).
S ome forthcoming and recent presentations :o Presenter, Recent findings in solar valuation, Center for Research into Regulated Industries
(Rutgers University) 27th Anrual Westem Conference, Monterey, California, Jwe,20l4.o Panelist, "Crossing the Analytical Chasm: Moving Energy Storage Modeling from Theory
to Practice, ESA 24th Annual Conference, Washington, DC, June 4-5,201.4.o Workshop Presenter, "Energy Storage 201," ESA 24th Annual Conference, Washington, DC,
Jvne 4-5,20L4.r Panelist, "Status of Large-Scale Solar Projects," Platts 9th Annual California Power
Markets, San Francisco, November 2013.
Managing Director, Economic and Pricing Analysis, BrightSource Energy, Oakland, CA
(llzOlt - 412013). Focus on the quantitative analysis of benefits of concentrating solar power
(CSP) with thermal energy storage, along with corresponding state and federal regulatory and
policy dimensions. Areas of analysis include valuation of energy, ancillary services, capacity, and
integration costs.o Lead author for CSP Alliance paper, "Economic and Reliability Benefits of CSP with
Thermal Energy Storage," August 2014, avulable at http://www.csp-al1iance.org/.o Membership on advisory groups for system modeling:
E. Udi Helman
October 2014
EXHIBIT 4OI
Page I of5
r Policy Advisory Committee, NREL project on concentrating solar power with
thermal storage with representatives from SCE, PG&E, CPUC, CEC, CAISO,
SMIID, LADWP, and other entities, August 2011 to present. Organized industry
sub-group to provide specific information on technology characteristics.. Policy Advisory Commiffee, Califomia Energy Commission/DNV-Gl project on
modeling concentrating solar power with therrnal storage, August 2012 to present.. CAISO 33% RPS renewable integration advisory team, 2011 - present.
o Directed economic benefits analysis in commercial negotiations with California IOU and
municipal utilities.o Extensive dialogue with state regulators in Califonria in support of economic valuation
findings. Many public presentations at state agencies, as listed below.o Primay author or lead reviewer/editor of numerous regulatory filings for BrightSource,
Large-Scale Solar Association (LSA) and other collaborators before CAISO, CPUC and
FERC (listed below).. Public presentations over 20ll-12 at Califonria Public Utilities Commission (CPUC),
Califomia Energy Qsmmission (CEC), Califorria ISO (CAISO), Federal Energy Regulatory
Commission (FERC), Departnent of Energy (DOE), Solar Power Intemational, Stanford
University, U.C. Berkeley POWER conference 2011, Center for Research into Regulated
lndustries (Rutgers University) annual Westem conference, Westem Independent
Transmission Group, Westem Resource Planners, and industry trade conferences by EUCI,
Platts, Infocast, and Marcus Evans.
Principal, Division of Markets & Infrastructure Development, California Independent System
Operator (CAISO), Folsom, CA (8/2007 to l/2011). Wide range ofprojects and roles:
Re n ew able inte gr atio n stu die s
Co-auttror of CAISO, "Integration of Renewable Resources: Operational Requirements and
Generation Fleet Capability at 200lo RPS," August 31,2010.
Numerous presentations on renewable integration analysis to CAISO stakeholders and
external audiences, including many at CPUC and CEC workshops; Panelist, FERC
Technical Conference,Integrating Renewable Resources into the Wholesale Electric Grid
(AD09-4-000), Washington, DC, March 2,2009.
Renewable transmission policy and other transmission projects
CAISO representative on Califomia Transmission Policy Group (CTPG) scenario team,
2010; Lead on CAISO team on the development of the renewable energy transmission
planning process, 2009-2010; Participant on CAISO team on the C3ETP transmission
proposal by PG&E; development of cost-benefit analysis and valuation of Helms pumped
storage plant upgrades.
Numerous presentations on reform of transmission policy to CAISO stakeholders and
extemal audiences, including: "Resource and transmission planning to achieve a 33% RPS
in Califomia - ISO modeling tools and planning framework," FERC Technical Conference
on Planning Models and Software, Washington, DC, June 10,2010.
Wholesale market design to support renewable integrationo Team lead for wholesale market design changes needed to address renewable integration.
Lead author on Califonria ISO, "Discussion Paper - Renewable lntegration: Market and
Product Review," July 8,2010.
E. Udi Helman
May2014
EXHIBIT 401
Page 2 of5
o Lead author, "Comments of the Califomia Independent System Operator Corporation,"
Integration of Variable Energy Resources, FERC Docket No. RM10-1 1-000, Apil12,2010.o Lead author on 250 page report on ISO and RTO progress towards integration of VERs,
http://www.isorro.org/site/cjhKoIZPBImE/b.4344503/k.83C1/FERC_Filings.htrn.o Organrzed2-day workshop on Wind Integration and ISOiRTO Wholesale Markets for the
ISO/RTO Council, Boston, June 8-9,2009.
Other roleso Lead on pricing of ISO backstop procurement of capacity from generators (equivalent to
Resource Adequacy contracts), 2007-2008 and 2010 onwards; prepared review of pricing
options and justification for tariff-based pricing rules.o CAISO representative on electricity sector team for California inter-agency Climate Action
Team, August 2007-present; point person for ISO perspective on GHG regulation,
including point of regulation design topics.o CAISO representative to Westem Climate Initiative, 2009-2010. Technical Advisory Group
of the Western Climate Initiative Electricity Subcommittee, 2008-2009.o CAISO representative and contributor, inter-agency Clean California Energy Future
initiative (CAISO, CPUC, CARB, CEC, CaIEPA),2009-2010. Final report on
implementation posted at
http://www.cacleanenergyfuture.ore/documents/CCEFImplementationPlan.pdf.
Economist, Division of Policy Analysis and Rulemaking, Office of Energy Markets and
Reliability, Federal Energy Regulatory Commission (FERC), Washington, DC (full time,
lll999 to 812007; part time consultant, llll997 to l/1999). Selected projects listed below.
Long-term Transmission Righ* in Organized Markets. Team lead for long-tenn transmission
rights in organized markets (L1,12006 to 8/2007); team memb er (3/2005 to 1112006). Coordinator
of compliance filings by RTOs and ISOs, with review responsibility for all filings and lead
responsibility for Midwest ISO. Primary author of StaffPaper on Long-term Transmission Rights
in Organized Markets (May 11,2005, Docket No. AD05-7-000).
Midwest ISO Market Design and Market Start-up. Lead economist from late 2002 to 2007 for
much of the analysis of Midwest Independent System Operator (ISO) market rules and start-up of
the Day 2 market. Worked on the major order that approved the Midwest ISO market design,
Midwest Independent Transmission System Operator, Inc.,108 FERC fl 61,163 (TEMT II Order)
(0812004\ and follow-up. Lead design of market-start safeguards which included several innovative
regulatory methods that were employed to enhance market operations in the first few months and
years. From 2005 to 2007, worked on a variety of orders addressing MISO market design issues,
including issues in marginal loss surplus refunds, redesign of the FTR rules and introduction of
long-term auction revenue rights (ARRs), and the design for a bid-based market for regulation and
operating reserves and scarcity pricing.
ISO-New England Market Design and Market Start-up. Lead economist from 1998 to 2002 for
much of the analysis of ISO-New England market rules. Wrote initial economic policy
memorandum on interim New England market rules in support of the FERC Order approving the
market design and market start (85 FERC \61,379; 12/17/1998).
E. Udi Helman
May2014
EXHIBIT 4OI
Page 3 of5
EDUCATION
Ph.D. Applied Microeconomics and Systems Analysis
Departneent of Geography and Environmental Engineering
The Johns Hopkins University, Balt'more, Maryland (2003)
PhD advisors: Benjamin Hobbs and John Boland
M.A. Environmental Studies
Institute of Environmental Studies/Deparhnent of Political Science
University of Toronto, Toronto, Canada (1989)
B.A. Political Science and Biology (minor)
University of Toronto, Toronto, Canada (1987)
SELECTED PI]BLICATIONS AND REPORTS
Helman, Udi, Economic and Reliability Benefits of Large-Scale Solar Plants," Chapter in
Lawrence fones (ed.), Renewable Energy lntegration: Practical Management of Variability,
Uncertainty and Flexibility in Power Grids, Elsevier, 2014.
Helman, Udi, and David Jacobowitz, "The Economic and Reliability Benefits of CSP with Thermal
Energy Storage: Recent Studies and Research Needs," CSP Alliance Report, August 2014,
available at http ://www.csp-alliance. org/.
Rothleder, Mark, Helman, Udi, Clyde foutan, Tao Guo, June Xie, and Sundar Venkataraman,
"Integration of wind and solar under a20Yo RPS: Stochastic simulation methods and results from
Califonria ISO studies," Proceedings of the IEEE, Power Engineering Society, 20l2lEEE Power &
Energy Society General Meeting, July 2217, 2012, P aper #20 I2GN/L892.
(co-author) Califonria ISO,Integration of Renewable Resources: Operational Requirements
and Generation Fleet Capability at2\o/o RPS, August31.,2010, available at
http : //www. caiso. com/Documents/Inte Eration-RenewableResources -
OperationalRequirementsandGenerationFleetCapabilityAt2OPercRP S,pdf.
Helman, Udi and Benjamin F. Hobbs, "Large-Scale Market Power Modeling: Analysis of the U.S.
Eastem Interconnection and Regulatory Applications," IEEE Transactions on Power Systems,
25, 3, 2010, pp. 1434-1448.
(co-author) ISO/RTO Council White Paper,'Variable Energy Resources, System Operations and
Wholesale Markets, Incorporating a Response to the Federal Energy Regulatory Commission's
Notice of Inquiry on Integration of Variable Energy Resources, Docket No. RM10-11-000,'April
12,2010, http://www.isorto.ors/site/c.ihKQzPBImE/b.43445031k.83C1/FERC_Filinss.htm.
Helman, Udi, Harry Singh, and Paul Sotkiewicz, "RTOs, Regional Electricity Markets, and Climate
Policy," in F. P. Sioshansi, ed., Generating Electricity in a Carbon Constrained World, Academic
Press,2009.
E. Udi Helman
May2014
EXHIBIT 4OI
Page 4 of5
Helman, Udi, Benjamin F. Hobbs, and Richard P. O'Neill, "The Design of US Wholesale Energy
and Ancillary Service Auction Markets: Theory and Practice," h F.P. Sioshansi, ed. Competitive
El ectricity Markets : D esign, Implementation, P erformance, Elsevier, 200 8.
O'Neill, Richard P., Udi Helman, Benjamin F. Hobbs, and Ross Baldick, "Independent System
Operators in the USA: History, Lessons Learned, and Prospects," in F. Sioshansi and W.
Pfaffenberger, Electricity Market Reform: An International Perspective, Elsevier, 2006.
Helman, Udi, "Market power monitoring and mitigation in the US wholesale power malkets,"
Energt, 3l (6-7), pp. 877 -904, 2006.
Baldick, Ross, Udi Helman, Benjamin F. Hobbs, and Richard P. O'Neill, "Design of Efficient
Generation Markets, Proceedings of the IEEE, Special Issue on Electric Power Systems:
Engineering and Policy, 93(11), November 2005.
O'Neill, Richard P., Ross Baldick, Udi Helman, Michael H. Rothkopf, and William Stewart,
"Dispatchable Transmission in RTO Markets," IEEE Transactions on Power Systems,2},l,2005,
l7l-179. IIEEE PES Technical Committee Prize Paper Award]
Hobbs, Benjamin, Udi HeLnan, Suradet Jitprapaikulsarn, Sreenivas Konda, and Dominic
Maratukulam, "Artificial neural networks for short-term energy forecasting: Accuracy and
economic value," Neuroc omputing, 23 ( 1 998), pp. 7 l -84.
Helman, Udi, Benjamin Hobbs, and John Boland. "Fishery Resource Values Used for Damage
Compensation in Maryland: Assessment of Need for Revisions." Report to the Power Plant
Research Program, Maryland Deparfinent of Natural Resources. Under Contract No. PR97-056-
001. January 4,1997.
U.S. Congress Office of Technology Assessment, Fueling Reform: Energlt Technologies for the
Former East Bloc. OTA-ETI-599 (Washington, DC: U.S. Govemment Printing Office, July 1994).
(Udi Helman and Joy Dunkerely co-authored Ch.7)
U.S. Congress Office of Technology Assessmert, Energt Efficiency Technologies for Central and
Eastern Europe. OTA-E-562 (Washington, DC: U.S. Government Printing Office, May 1993).
(authorofCh.5)
E. Udi Helman
May2014
EXHIBIT 401
Page 5 of5
I hereby certiff that on *" fiuyof October ,2Ol4,I caused to be served, viathe
method(s) indicated below, true and correct copies of the foregoing document, upon:
Jean Jewell, Secretary
Idaho Public Utilities Commission
472 W est Washinglon Steet
P.O. Box 83720
Boise,ID 83720-0074
ijewell@puc. state.id.us
Kristine Sasser, Deputy Attorney General
Idaho Public Utilities Commission
472W. Washington St.
Boise,Idaho 83720
kris.sasser@nuc. idaho. gov
Donovan E. Walker
Greg Said
Michael f. Youngblood
Regulatory Dockets
Idaho Power Company
1221 West Idatro Street
P.O. Box 70
Boise,ID 83707
dwalker@idahopower. com
[e-mail]
myoun gblood@ idahopower. com
dockets@ idahopower. com
Idaho Conservation League
c/o Benjamin J. Otto
710 N. 6th St.
Boise,Idaho 83702
botto @ idahoconservation. org
KenMiller
Snake River Alliance
P.O. Box l73l
Boise,ID 83701
kmiller@snakeriveralliance. ore
CERTIFICATE OF SERVICE
Hand Delivered
U.S. Mail
Fa,x
Fed. Express
Email
Hand Delivered
U.S. Mail
Fa"x
Fed. Express
Email
Hand Delivered
U.S. Mail
Fax
Fed. Express
Email
Hand Delivered
U.S. Mail
Fax
Fed. Express
Email
Hand Delivered
U.S. Mail
Fax
Fed. Express
Email
X
(J
&(l
(.J
(J
e(J
I
t{
i(J
(J
E
(J
tJv
tl
LJ
x
irr & Mu,r,ER LLP