20211001Ellsworth Direct.pdf

BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION

IN THE MATTER OF IDAHO POWER
COMPANY’S APPLICATION FOR
APPROVAL TO MODIFY ITS DEMAND
RESPONSE PROGRAMS.

)
)
)
)
)
)
)

CASE NO. IPC-E-21-32

IDAHO POWER COMPANY

DIRECT TESTIMONY

OF

JARED L. ELLSWORTH
ELLSWORTH, DI 1
Idaho Power Company
Q. Please state your name, business address, and 1
present position with Idaho Power Company (“Idaho Power” or 2
“Company”). 3
A. My name is Jared L. Ellsworth and my business 4
address is 1221 West Idaho Street, Boise, Idaho 83702. I am 5
employed by Idaho Power as the Transmission, Distribution & 6
Resource Planning Director for the Planning, Engineering & 7
Construction Department. 8
Q. Please describe your educational background. 9
A. I graduated in 2004 and 2010 from the 10
University of Idaho in Moscow, Idaho, receiving a Bachelor 11
of Science Degree and Master of Engineering Degree in 12
Electrical Engineering, respectively. I am a licensed 13
professional engineer in the State of Idaho. 14
Q. Please describe your work experience with 15
Idaho Power. 16
A. In 2004, I was hired as a Distribution 17
Planning engineer in the Company’s Delivery Planning 18
department. In 2007, I moved into the System Planning 19
department, where my principal responsibilities included 20
planning for bulk high-voltage transmission and substation 21
projects, generation interconnection projects, and North 22
American Electric Reliability Corporation’s (“NERC”) 23
reliability compliance standards. I transitioned into the 24
Transmission Policy & Development group with a similar 25
ELLSWORTH, DI 2
Idaho Power Company
role, and in 2013, I spent a year cross-training with the 1
Company’s Load Serving Operations group. In 2014, I was 2
promoted to Engineering Leader of the Transmission Policy & 3
Development department and assumed leadership of the System 4
Planning group in 2018. In early 2020, I was promoted into 5
my current role as the Transmission, Distribution and 6
Resource Planning Director. I am currently responsible for 7
the planning of the Company’s wires and resources to 8
continue to provide customers with cost-effective and 9
reliable electrical service. 10
Q. What is the purpose of your testimony in this 11
case? 12
A. The purpose of my testimony is to (1) inform 13
the Idaho Public Utility Commission (“Commission”)of recent 14
improvements in the determination of the capacity 15
contribution of supply-side and demand-side resources, 16
developed as part of the 2021 Integrated Resource Plan 17
(“IRP”) planning process, and how such improvements have 18
impacted the evaluation of Idaho Power’s Demand Response 19
(“DR”) portfolio, (2) explain the analysis used to identify 20
the proposed DR portfolio operating parameters using an 21
enhanced risk-based methodology, and (3) describe how the 22
economic value of the DR portfolio was determined and how 23
that economic value was used to inform the program 24
compensation modifications presented in this case. 25
ELLSWORTH, DI 3
Idaho Power Company
I. IRP CHANGES IMPACTING DEMAND RESPONSE EVALUATION 1
Q. Generally, how has DR been considered as a 2
resource in the Company’s planning process? 3
A. Historically, the Company has evaluated the 4
maximum operational potential of its existing DR resources 5
by their ability to meet the peak demand hour (peak load) 6
during the summer months of June through August throughout 7
the IRP planning horizon. This is consistent with how 8
traditional supply-side resources have been evaluated. 9
Q. How did the methodology used to analyze the 10
Company’s capacity value of DR change following the 2019 11
IRP? 12
A. When determining the capacity value of the 13
Company’s DR portfolio in the 2019 IRP, the calculation was 14
based on the DR portfolio’s ability to be utilized during 15
the top one-hundred system load hours given the program 16
parameters. 17
Moving into the 2021 IRP planning process, the 18
Company adopted a risk-based methodology, known as 19
Effective Load Carrying Capability (“ELCC”), to evaluate 20
the capacity contribution of the Company’s existing 21
resources, expected future resources (including variable 22
resources), and DR. This method evaluates the Company’s 23
load and resource balance at the time of the highest-risk 24
hours, rather than only analyzing a resource’s ability to 25
ELLSWORTH, DI 4
Idaho Power Company
meet peak load. 1
Q. Why is the Company proposing to evaluate its 2
load and resource balance using the ELCC risk-based 3
methodology rather than the capacity planning method based 4
on system peak load? 5
A. As previously mentioned, Idaho Power’s 6
planning process historically focused on ensuring adequate 7
resources were available to meet the overall system peak 8
load. Due to the penetration of solar, wind, and other 9
variable resources connected to the Idaho Power system, the 10
primary hours of need for additional resources, or the 11
highest-risk Loss-of-Load Probability (“LOLP”) hours, are 12
no longer expected to align with the hours of Idaho Power’s 13
system peak load. 14
Q. What is LOLP? 15
A. LOLP is the statistical likelihood, between 16
zero and one, of the system demand exceeding the available 17
generating capacity during a given time period, typically 18
an hour. The LOLP for an hour can be calculated by 19
comparing the system net load to a statistically derived 20
resource capacity probability distribution curve for any 21
given hour. The resource capacity probability distribution 22
curve is the probability (based on resource capacities, 23
historical resource availability, and statistical forced 24
outage rates) the Company will have more than a certain 25
ELLSWORTH, DI 5
Idaho Power Company
amount of generation available to it at any given time. 1
Q. What is net load? 2
A. Net load is the total system load minus any 3
non-controllable resources, i.e., generation that is either 4
(1) not controlled by Idaho Power, or (2) has limited or 5
zero flexibility. Examples of generation resources Idaho 6
Power does not have operational control over are wind, 7
solar, and PURPA resources. Run-of-river hydro is an 8
example of a resource with limited flexibility. 9
Q. How do the highest-risk LOLP hours compare to 10
the time of the system peak load? 11
A. In the preliminary 2021 IRP analyses, the 12
highest-risk LOLP hours have been shown to shift to later 13
in the day when solar sees an output reduction. As more 14
solar comes on to the Company’s system, the Company expects 15
the LOLP of the evening solar-ramping-hours to increase and 16
drive the need for additional resources later in the day. 17
Charts 1 through 4 depict Idaho Power’s LOLP hours 18
under various solar resource scenarios and how the highest-19
risk hours begin to shift as more solar is added to the 20
system. While the time of the Company’s system peak load 21
has historically occurred between 5:00pm and 8:00pm, the 22
highest-risk hours are expected to occur between 7:00pm and 23
10:00pm, with some medium-risk hours leading up to 7:00pm 24
and from 10:00pm to 11:00pm, over the 2021 IRP planning 25
ELLSWORTH, DI 6
Idaho Power Company
horizon. 1
Chart 1 reflects July’s hourly LOLP with no solar 2
resources. The highest-risk hours are between 3:00pm and 3
8:00pm. 4
Chart 1. July LOLP - Test Year 2 Shape – No Solar Resources 5
and No Demand Response 6
7
// 8
// 9
// 10
// 11
// 12
// 13
// 14
ELLSWORTH, DI 7
Idaho Power Company
Chart 2 reflects July’s hourly LOLP with 316 1
Megawatts (“MW”) of solar resources on the system, which is 2
reflective of the current solar capacity in 2020. The 3
highest-risk hours of need shift later in the day from 4
4:00pm to 9:00pm. 5
Chart 2. July LOLP - Test Year 2 Shape – 2020 Solar 6
Resources and No Demand Response 7
8
// 9
// 10
// 11
// 12
// 13
// 14
// 15
ELLSWORTH, DI 8
Idaho Power Company
Chart 3 reflects July’s hourly LOLP with 436 MW of 1
solar resources on the system, which includes the 120 MW 2
expected from the Jackpot Solar Project in 2023. The 3
highest-risk hours continue to shift later in the day, 4
moving to 5:00pm to 10:00pm. This scenario also shows the 5
10:00pm and 11:00pm hours starting to have a higher risk 6
probability. 7
Chart 3. July LOLP - Test Year 2 Shape – 2023 Solar 8
Resources and No Demand Response 9
10
// 11
// 12
// 13
// 14
// 15
ELLSWORTH, DI 9
Idaho Power Company
Finally, Chart 4 reflects July’s hourly LOLP with 836 1
MW of solar resources on the system, which includes 400 2
additional MW as compared to the connected 2023 solar 3
capacity and is potentially reflective of the future system 4
as solar becomes more prevalent. The highest-risk hours 5
really condense into the later hours of the day. 6
Chart 4. July LOLP - Test Year 2 Shape – Future Solar and 7
No Demand Response 8
9
// 10
// 11
// 12
// 13
// 14
// 15
ELLSWORTH, DI 10
Idaho Power Company
Q. Why does the Company believe the ELCC method 1
is appropriate for resource planning purposes? 2
A. The assumption that the highest-risk hours of 3
capacity shortfall directly correspond with the hours of 4
highest load is only valid for a system with little or no 5
variable resource penetration. With the Company’s existing 6
resources, and the projected additions of more variable 7
resources coming onto the system, the hours of highest-risk 8
will not necessarily align with the hours of highest load. 9
The Company believes that the ELCC method accurately 10
captures the Company’s future resource adequacy risks. The 11
ELCC method still considers DR’s ability to contribute 12
capacity given the program parameters, but the hours of 13
need are identified using probabilistic and statistical 14
methods as opposed to utilizing the top one-hundred system 15
load hours. 16
Q. How is ELCC applied in the resource planning 17
process? 18
A. ELCC is a reliability-based metric used to 19
determine the peak capacity credit of any given resource 20
and captures an individual generator’s contribution to 21
overall system reliability. It is primarily driven by the 22
timing of high-risk LOLP hours. For example, a generator 23
that contributes a significant level of capacity during 24
high-risk LOLP hours will have a higher ELCC than a 25
ELLSWORTH, DI 11
Idaho Power Company
resource that delivers the same capacity during medium to 1
low-risk LOLP hours. Utilizing multiple test years, ELCC 2
values are determined and assigned to existing and 3
selectable resources in the Aurora model for different 4
scenarios, sensitivities, and portfolios in the IRP. 5
Q. How is ELCC used to calculate the capacity 6
contribution of various resources? 7
A. The ELCC of a resource is determined through a 8
multi-step process. First, the Company calculates the 9
perfect generation, in MW, required for the system to 10
achieve a Loss-of-Load Expectation (“LOLE”) of 0.05 days 11
per year with all market purchases set equal to zero. An 12
LOLE of 0.05 days per year represents the statistical 13
probability that the Company’s available generation 14
capacity is only insufficient to serve demand one time in 15
the span of twenty years. Next, the resource being 16
evaluated is added to the system and the Company once again 17
calculates the perfect generation required to meet the same 18
LOLE threshold. The perfect generation of the system with 19
the resource is subtracted from the perfect generation of 20
the system without the resource and then divided by the 21
evaluated resource’s nameplate capacity to obtain the 22
resource’s ELCC as shown in the equation below. 23
// 24
// 25
ELLSWORTH, DI 12
Idaho Power Company
𝐄𝐋𝐂𝐂 𝐏𝐆𝟏 𝐏𝐆𝟐
𝐑𝐞𝐬𝐨𝐮𝐫𝐜𝐞𝐍𝐏
1
2
Q. What is the difference between LOLP and LOLE? 3
A. LOLP identifies high and low-risk hours in 4
regards to system load exceeding generation capacity, and 5
the maximum LOLP from each day over the course of 365 days 6
are summed together to calculate the LOLE. 7
Q. What is an Equivalent Forced Outage Rate 8
(“EFOR”) and how is it incorporated into the ELCC 9
methodology? 10
A. An EFOR represents the number of hours a 11
generation unit is forced off-line compared to the number 12
of hours the unit runs (planned maintenance is not factored 13
into a unit's EFOR). For example, an EFOR of 3 percent 14
means a generator is forced off, or incurs an unplanned 15
outage, 3 percent of its running time. A perfect generator 16
is a generation unit whose EFOR value is 0 percent, meaning 17
that it is always available and never forced off-line. A 18
perfect resource does not actually exist in practice, but 19
it is used as a “standard” to enable ease of comparison 20
between different generation options. 21
Q. What LOLE threshold is the Company planning to 22
utilize in the 2021 IRP? 23
𝐏𝐆𝟏 𝐏𝐞𝐫𝐟𝐞𝐜𝐭 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 𝐫𝐞𝐪𝐮𝐢𝐫𝐞𝐝 𝐭𝐨 𝐚𝐜𝐡𝐢𝐞𝐯𝐞 𝐋𝐎𝐋𝐄 𝐨𝐟 𝟎.𝟎𝟓 𝐝𝐚𝐲𝐬/𝐲𝐞𝐚𝐫 𝐰𝐢𝐭𝐡𝐨𝐮𝐭 𝐢𝐧𝐜𝐥𝐮𝐝𝐢𝐧𝐠 𝐞𝐯𝐚𝐥𝐮𝐚𝐭𝐞𝐝 𝐫𝐞𝐬𝐨𝐮𝐫𝐜𝐞
𝐏𝐆𝟐 𝐏𝐞𝐫𝐟𝐞𝐜𝐭 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 𝐫𝐞𝐪𝐮𝐢𝐫𝐞𝐝 𝐭𝐨 𝐚𝐜𝐡𝐢𝐞𝐯𝐞 𝐋𝐎𝐋𝐄 𝐨𝐟 𝟎.𝟎𝟓 𝐝𝐚𝐲𝐬/𝐲𝐞𝐚𝐫 𝐰𝐡𝐞𝐧 𝐢𝐧𝐜𝐥𝐮𝐝𝐢𝐧𝐠 𝐞𝐯𝐚𝐥𝐮𝐚𝐭𝐞𝐝 𝐫𝐞𝐬𝐨𝐮𝐫𝐜𝐞
𝐑𝐞𝐬𝐨𝐮𝐫𝐜𝐞𝐍𝐏 𝐍𝐚𝐦𝐞𝐩𝐥𝐚𝐭𝐞 𝐜𝐚𝐩𝐚𝐜𝐢𝐭𝐲 𝐨𝐟 𝐭𝐡𝐞 𝐞𝐯𝐚𝐥𝐮𝐚𝐭𝐞𝐝 𝐫𝐞𝐬𝐨𝐮𝐫𝐜𝐞
ELLSWORTH, DI 13
Idaho Power Company
A. In the 2021 IRP, the Company will plan for an 1
LOLE threshold of 0.05 days per year, which represents a 2
statistical probability of the Company being resource 3
insufficient one day in twenty years. The two primary 4
reasons the Company ultimately chose to plan for a 0.05 5
days per year LOLE threshold are: 6
(1) The Company operates a power system highly 7
dependent on hydroelectric resources, which can vary 8
dramatically from year-to-year due to water conditions. A 9
poor water year can significantly affect hydroelectric 10
resource availability, especially during summer months. 11
(2) The LOLE methodology utilizes historical data to 12
make forward looking decisions. Recently, weather extremes 13
have been occurring with greater frequency. Therefore, by 14
planning for an LOLE of 0.05 days per year, the Company 15
expects to be able to maintain a similar level of 16
reliability that Idaho Power’s customers and regulators 17
expect moving forward. 18
Q. How are different resource types modeled in 19
the LOLE calculation? 20
A. Idaho Power’s resources are split into three 21
primary categories: dispatchable resources, non-22
controllable resources, and energy-limited resources. 23
Idaho Power’s dispatchable resources include the 24
Hells Canyon Complex, natural gas plants, Bridger and Valmy 25
ELLSWORTH, DI 14
Idaho Power Company
coal plants, and various transmission assets. These 1
resources are modeled using a monthly outage table that 2
factors in their generating capacities and EFORs. 3
Non-controllable resources are modeled by using four 4
years of historical hourly output data to provide realistic 5
weather shapes. Non-controllable resources are resources for 6
which Idaho Power does not have direct operational control 7
over such as wind, solar, dairy digestors, non-wind and 8
non-solar PURPA projects, run-of-river hydroelectric 9
plants, and geothermal generation. 10
Dispatch shapes for energy-limited resources such as 11
battery storage and DR are created based on net load 12
explained later in my testimony. 13
Q. What did the Company find when evaluating its 14
existing DR programs utilizing the ELCC method? 15
A. The existing DR programs, as structured, are not 16
effective at meeting system needs over the planning 17
horizon. As more fully described below, certain parameters 18
of the existing programs, specifically the current dispatch 19
hours and the program season, limited the effectiveness of 20
DR as a resource. 21
II. PROPOSED DEMAND RESPONSE PROGRAM PARAMETERS 22
Q. How were the existing DR programs evaluated 23
using the LOLP and ELCC methodology? 24
ELLSWORTH, DI 15
Idaho Power Company
A. Idaho Power conducted the ELCC analysis on the 1
current DR programs to identify how effective they are at 2
meeting future high-risk LOLP hours. 3
The ELCC of DR was calculated using a multi-step 4
process. First, every day in a test year was sorted from 5
highest to lowest based on their net peak load in MW. 6
Second, a daily MW target was set for each day based on the 7
highest net load hour within the day and the size of the 8
dispatchable DR group. The Company determined that an 9
approximate 50 MW group size results in a capacity amount 10
that is operationally manageable yet still large enough to 11
have a meaningful impact on reducing system load. It also 12
most closely aligns with how Idaho Power’s Load Serving 13
Operations group dispatches the programs. 14
After sorting the days and establishing a daily 15
target, the analysis identified if the day was within the 16
DR season’s start and end dates and if the day was not a 17
weekend or holiday. If the day met the DR program 18
parameters, the algorithm would analyze each hour of the 19
day and compare the hourly net load with the daily target. 20
If the net load was above the target, the function would 21
dispatch DR MW groups in that hour. The algorithm then 22
iterates over the remaining hours in that day until DR had 23
been dispatched to reduce the net load for each of the 24
hours initially above the daily target, or DR capacity had 25
ELLSWORTH, DI 16
Idaho Power Company
been exhausted. After completing one day, the algorithm re-1
sorts all of the remaining days in the test year by net 2
peak load and repeats the process. In this manner, the 3
algorithm dispatches the DR programs in a way that 4
maximizes their usage and effectiveness. The algorithm 5
lastly creates a dispatch pattern by adding all the groups 6
into a single load shape. 7
The ELCC of DR is obtained by first determining the 8
perfect generation needed to achieve an LOLE of 0.05 days 9
per year without any DR on the system. Next, the DR load 10
shape derived using the algorithm described above is added 11
to the system, and the perfect generation is calculated 12
again. The ELCC of DR is then calculated by taking the 13
difference between the two perfect generation values and 14
dividing it by the DR portfolio’s nameplate capacity. 15
Q Describe the algorithm the Company utilized in 16
the analysis. 17
A. The Company internally developed the LOLE 18
MATLAB® algorithm within the MATLAB® software it is 19
utilizing for the IRP planning process. The algorithm is 20
computationally intensive, composed of several scripts, and 21
requires specialized software to effectively and accurately 22
run the calculations that model Idaho Power’s existing and 23
future resources. 24
// 25
ELLSWORTH, DI 17
Idaho Power Company
Q. What was the result of the ELCC analysis for 1
the Company’s existing DR programs? 2
A. Using the current program parameters, the ELCC 3
of a 380 MW DR portfolio is estimated to be approximately 4
17 percent. That is, of the total 380 MW DR portfolio 5
capacity, only 65 MW can be relied upon to meet the 6
highest-risk LOLP hours. The analysis was completed by 7
evaluating the current DR programs over four historical 8
test years. The historical test years were used for weather 9
shaping, scaled to have the same peak load as the 10
forecasted 2023 system peak, and included the known solar 11
resources that will be online in 2023 such as the 120 MW 12
Jackpot Solar Project. 13
Table 1 below shows the resulting ELCC value for 14
each test year and the average value across those years. 15
Based on this analysis, and given the current program 16
parameters, the current DR portfolio is expected to only be 17
17 percent effective in meeting high-risk LOLP hours. 18
Table 1. Current Demand Response Program ELCC 19
Test Year Current Demand Response ELCC (%) 
Test Year 1 14.21% 
Test Year 2 33.20% 
Test Year 3 12.10% 
Test Year 4 7.90% 
Average 16.84% 
20
ELLSWORTH, DI 18
Idaho Power Company
Q. How were the proposed DR program parameters 1
determined? 2
A. Recognizing that the existing program 3
parameters may limit the effectiveness of DR, the Company 4
conducted several sensitivity analyses to determine the 5
parameter adjustments needed to more effectively meet high-6
risk LOLP hours. These analyses were performed by modifying 7
several program criteria and evaluating the impact to the 8
ELCC of the DR portfolio. The program criteria studied for 9
each program included events per week, events per season, 10
time available, length of program season, and total hours 11
dispatched per week. 12
Q. What DR program parameters did the Company 13
conclude will more effectively meet future high-risk LOLP 14
hours? 15
A. The sensitivity analyses concluded that the 16
dispatch times available and the length of program season 17
were the two parameters that had the highest impact on the 18
ELCC of DR. Therefore, the proposed parameters that more 19
effectively meet future high-risk LOLP hours were 20
determined as outlined in Table 2 below. 21
// 22
// 23
// 24
// 25
ELLSWORTH, DI 19
Idaho Power Company
Table 2. Current and Proposed Demand Response Program 1
Parameters 2
Parameter Current Parameters Proposed Parameters 
Events per Week 15 hours 16 hours 
Events per Season 60 hours 60 hours 
Time Available 1:00pm to 9:00pm 3:00pm to 11:00pm 
Season Dates June 15th to August 15th  June 15th to September 15th  
Holidays No holidays No holidays 
3
Q. Did Idaho Power analyze the effectiveness of 4
varying levels of DR capacity? 5
A. Yes. As shown in Chart 5, the Company analyzed 6
the effectiveness of DR capacity in 50 MW increments. The 7
chart shows that DR effectiveness, and therefore ELCC, 8
diminishes as DR nameplate capacity increases. 9
Chart 5. DR Effectiveness vs DR Nameplate Capacity 10
11 12
// 13
0
50
100
150
200
250
300
350
50 100 150 200 250 300 350 400 442 492
DR
 Ef
f
e
c
t
i
v
e
n
e
s
s
 (M
W
)
DR Nameplate (MW)
DR Effectiveness
Test Year 1 Test Year 2 Test Year 3 Test Year 4 Average
ELLSWORTH, DI 20
Idaho Power Company
Q. What is the ELCC of the DR portfolio using the 1
proposed parameters? 2
A. The ELCC of DR is dependent on the nameplate 3
capacity of the program as shown in Chart 5. While the 4
nameplate of the proposed DR portfolio is still unknown, 5
the Company estimates the approximate ELCC of a DR 6
portfolio with the proposed parameters to be 56 percent 7
with a 380 MW nameplate capacity. This would be 8
approximately a 40 percent improvement in effectiveness 9
from the current program parameters. As mentioned earlier, 10
the main drivers of the increased effectiveness are the 11
shift in the dispatch time period and extending the program 12
season by one month to September 15th. 13
III. COST-EFFECTIVENESS COMPONENTS 14
Q. Please explain how the primary inputs to the 15
proposed economic evaluation described in Mr. Nesbitt’s 16
testimony were determined. 17
A. The three components of the proposed 18
alternative cost calculation are (1) the levelized capacity 19
fixed costs of a proxy resource, (2) the additional system 20
benefits of the proxy resource, and (3) the ELCC of the 21
annual DR nameplate capacity compared to a proxy resource. 22
Q. Describe how the value of the levelized 23
capacity fixed costs of a proxy resource was determined. 24
ELLSWORTH, DI 21
Idaho Power Company
A. The proxy resource used to evaluate the cost-1
effectiveness of the proposed DR programs was a Simple-2
Cycle Combustion Turbine (“SCCT”). Through the 2021 IRP 3
resource costing process, the 2022 levelized fixed cost 4
value was determined to be $131.60 per kW per year. This 5
value represents the fixed cost per kW per year if the 6
Company were to build the SCCT instead of running the DR 7
programs. 8
Q. Describe how a value for the additional system 9
benefits of the proxy resource is determined. 10
A. Because an SCCT is not restricted to operate 11
for only a defined number of hours like DR, it provides 12
additional benefits and reliability to the Company’s 13
system. To determine the approximate value of the 14
additional benefit, the Company completed a Production Cost 15
Model (“PCM”) in Aurora by including the SCCT and DR in two 16
separate runs of a 2019 IRP PCM subset. The additional 17
system benefits of the SCCT over the first 5 years of the 18
planning horizon equated to $38.11 per kW per year compared 19
to an equally effective 492 MW DR portfolio. The Company 20
believes calculating the additional benefit value is 21
important because it identifies the economic value a supply 22
side resource has beyond the system benefits DR can 23
provide. 24
// 25
ELLSWORTH, DI 22
Idaho Power Company
Q. Describe how the ELCC of the annual DR 1
nameplate capacity compared to a proxy resource is 2
determined. 3
A. The ELCC of the annual DR nameplate capacity 4
is calculated by first obtaining the amount of perfect 5
generation displaced by the proposed DR programs, i.e. DR’s 6
perfect resource effectiveness. An SCCT is not a perfect 7
generator and has an EFOR greater than zero. By using the 8
same ELCC methodology, but factoring in an SCCT’s EFOR, the 9
Company can determine the size of an effective-equivalent 10
SCCT to the DR portfolio’s perfect resource effectiveness. 11
The ELCCSCCT of the DR programs is calculated by taking the 12
quotient between the capacity of the effective-equivalent 13
SCCT and the DR nameplate capacity. Using a 492 MW DR 14
portfolio, the effective equivalent SCCT’s nameplate was 15
determined to be 272 MW based on the Company’s projected 16
2023 load and resource balance. This results in an ELCCSCCT 17
of 55 percent given a DR portfolio capacity of 492 MW (272 18
MW ÷ 492 MW = 55%). 19
Q. Why is the Company utilizing 492 MW as the 20
maximum nameplate DR in its economic evaluation analysis? 21
A. The Company utilized a Northwest Power and 22
Conservation Council (“NWPCC”) assessment1 of DR potential 23

1 2021 Northwest Power Plan Supporting Material – Demand Response.
Council Document 21-5 (September 2021) available at
https://www.nwcouncil.org/2021powerplan_demand-response
ELLSWORTH, DI 23
Idaho Power Company
for the Northwest region to determine the DR potential that 1
may be available in Idaho Power’s service area. The Company 2
concluded that Idaho Power’s service area has 584 MW of DR 3
potential as shown in Exhibit 1 to my testimony. From the 4
584 MW of potential, the Company determined that 492 MW 5
would have similar program parameters such as seasonal 6
restrictions, hours per year, etc. Therefore, 492 MW of 7
traditional nameplate DR can be modeled using the same ELCC 8
methodology, and the results can be utilized to establish 9
programs and incentives that will remain cost-effective as 10
the size of the DR portfolio changes over the planning 11
horizon. 12
Q. Will the Company perform any additional 13
analysis to validate the value of DR in the 2021 IRP? 14
A. Yes. The Company intends to perform an 15
updated analysis to validate that DR provides system value. 16
The Company intends to run a PCM in Aurora with the 17
proposed DR programs through the IRP planning horizon. This 18
PCM will be based on the preferred portfolio in the 19
Company’s 2021 IRP. A second PCM run will be performed that 20
will replace the proposed DR programs with an effective-21
equivalent SCCT. The total PCM cost over the time horizon 22
will be compared between the two runs. The extra benefit 23
that the SCCT provides (in dollars per kW per year) will be 24
calculated by taking the difference of the run with the 25
ELLSWORTH, DI 24
Idaho Power Company
effective-equivalent SCCT and the run with the proposed DR 1
portfolio each year and dividing it by the capacity of the 2
effective-equivalent SCCT. 3
Q. Has the Company determined how it will 4
evaluate existing and expanded DR in future IRPs? 5
A. Yes. Through the coordination of both this 6
filing and the upcoming 2021 IRP filing, the Company has 7
determined that for the 2021 IRP, it will evaluate DR in 8
the IRP modeling process by utilizing the 584 MW of DR 9
potential identified in the NWPCC assessment. 10
The 584 MW of DR potential will consider a 11
conservative estimate of 300 MW of capacity from the 12
modified DR programs. Therefore, a maximum of 13
approximately an additional 280 MW of DR (584–300 MW 14
rounded down) will be available for selection in the Aurora 15
model when analyzing the future load and resource balance. 16
The additional DR capacity will be divided into 20 MW 17
bundles and available for selection up to the 280 MW 18
threshold. 19
For future IRPs, the Company intends to evaluate the 20
possibility of conducting an Idaho Power specific potential 21
study to evaluate the DR potential in its service area. 22
Q. How did the Company determine the 300 MW of DR 23
capacity to be included as a committed resource in the 2021 24
IRP analysis? 25
ELLSWORTH, DI 25
Idaho Power Company
A. Because the Company’s DR programs are existing 1
resources that have been included in past IRP analyses, the 2
Company needed to determine a committed capacity value for 3
inclusion in the 2021 IRP model. As described in Mr. 4
Nesbitt’s testimony, the Company expects there will be an 5
impact to DR participation as a result of the proposed 6
parameter changes. Therefore, the Company took a 7
conservative approach by adjusting the current 380 MW 8
capacity of the DR programs to 300 MW. 9
If the DR portfolio was to have a nameplate capacity 10
of 300 MW, the ELCC of the DR portfolio would increase to 11
63 percent from 56 percent with a 380 MW DR portfolio. 12
Table 3 below shows the ELCC for each test year and the 13
average across those years for a DR portfolio with the 14
proposed parameters and a capacity of 300 MW. 15
Table 3. Proposed Demand Response Portfolio ELCC (300 MW 16
Capacity) 17
Test Year Proposed Demand Response ELCC (%) 
Test Year 1 51.33% 
Test Year 2 64.00% 
Test Year 3 62.00% 
Test Year 4 74.33% 
Average 62.92% 
18
A 300 MW DR portfolio will result in a 189 MW 19
capacity contribution to the 2022 load and resource balance 20
(300 MW x 62.92% = 189 MW). 21
ELLSWORTH, DI 26
Idaho Power Company
Q. How will the Company proceed if the IRP 1
selects the 20 MW bundles of additional DR for the future 2
load and resource balance? 3
A. If the 2021 IRP model selects additional 20 MW 4
DR bundles in future years, the Company will work 5
internally, and with the Energy Efficiency Advisory Group, 6
to determine the best course of action on DR expansion. The 7
Company will evaluate potential options such as increasing 8
the capacity of current DR programs or adding new programs 9
to the Company’s DR portfolio. 10
IV. CONCLUSION 11
Q. Please summarize your testimony. 12
A. The Company has applied an enhanced risk-based 13
capacity planning methodology to determine new proposed 14
operating parameters that will greatly improve the ability 15
of DR to meet future high-risk LOLP hours. The Company then 16
built upon this new planning methodology to develop a 17
proposed economic valuation calculation to inform program 18
compensation levels and cost-effectiveness. The Company 19
believes that the enhanced analysis has identified changes 20
that improve the DR programs to the benefit of customers, 21
the system, and meeting future resource needs. 22
Q. Does this complete your testimony? 23
A. Yes, it does. 24
// 25
ELLSWORTH, DI 27
Idaho Power Company
DECLARATION OF JARED L. ELLSWORTH 1
I, Jared L. Ellsworth, declare under penalty of 2
perjury under the laws of the state of Idaho: 3
1. My name is Jared L. Ellsworth. I am employed 4
by Idaho Power Company as the Transmission, Distribution & 5
Resource Planning Director for the Planning, Engineering & 6
Construction Department. 7
2. On behalf of Idaho Power, I present this 8
pre-filed direct testimony and Exhibit No. 1 in this 9
matter. 10
3. To the best of my knowledge, my pre-filed 11
direct testimony and exhibit are true and accurate. 12
I hereby declare that the above statement is true to 13
the best of my knowledge and belief, and that I understand 14
it is made for use as evidence before the Idaho Public 15
Utilities Commission and is subject to penalty for perjury. 16
SIGNED this 1st day of October 2021, at Boise, 17
Idaho. 18
19
Signed: 20
21
22
23
24
25
IPC Demand Response Potential Analysis Based on NWPCC's Draft 2021 Northwest Power Plan (September 2021)
Product
2041 NWPCC Ramped 
Achievable Potential 
MW by Year by Product
Idaho Power MW 
Allocation1 Product Summer Achievable 
Potential (MW)
Idaho Power MW 
Allocation3
IPC DR Potential 
Allocation
NWPPC Region 
(MW)
IPC Service Area 
(MW)
IPC 
Allocation 
Ratio (%4
ResCPP 251.4 29.9 ResCPP 251.4 29.9 System Peak 31,125 3,700 11.89%
ComCPP 134.1 15.9 ComCPP 134.1 15.9 DR Potential 2,386 284 n/a
IndCPP 108.7 12.9 IndCPP 108.7 12.9
IndRTP 24.2 2.9 IndRTP 24.2 2.9
ResERWHDLCGrd 867.3 103.1 ResERWHDLCGrd 867.3 103.1 Potential (MW)
ResERWHDLCSwch 76.5 9.1 ResERWHDLCSwch 76.5 9.1 IPC Estimated Minimum DR Capacity with Proposed Changes 300
ResBYOT 63.1 7.5 ResBYOT 63.1 7.5 IPC Service Area DR Potential + 284
NRCoolSwchMed 48.1 5.7 NRCoolSwchMed 48.1 5.7 IPC DR Total Potential 584
NRTstatSm 23.0 2.7 NRTstatSm 23.0 2.7 Possible Future Programs with Different ELCC ‐92
NRCoolSwchSm 15.2 1.8 NRCoolSwchSm 15.2 1.8 IPC Traditional DR Potential 492
NRIrrSmMed 464.0 55.2 NRIrrSmMed 0.0 0.0
NRIrrLg 393.7 46.8 NRIrrLg 0.0 0.0
ResACSwch 194.1 23.1 ResACSwch 0.0 0.0
NRCurtailInd 174.1 20.7 NRCurtailInd 0.0 0.0
NRCurtailCom 38.7 4.6 NRCurtailCom 0.0 0.0
DVR 560.9 66.7 DVR 560.9 66.7
ResTOU 214.0 25.4 ResTOU 214.0 25.4
ResEVSEDLCSwch 72.3 8.6 ResEVSEDLCSwch 0.0 0.0
ResHPWHDLCGrd 6.3 0.7 ResHPWHDLCGrd 0.0 0.0
ResHPWHDLCSwch 0.8 0.1 ResHPWHDLCSwch 0.0 0.0
MW Total 3,73 443 MW Total 2,386 284
(1) Multiplied by 11.89% allocation ratio
(2) Removes potential from current IPC programs and programs considered to be high cost
(3) Multiplied by 11.89% allocation ratio
(4) Calculated by taking IPC's proportion of the region's system peak (3,700 ÷ 31,125 = 11.89%)
(5) DVR + ResTOU = Future Programs with Different ELCC (66.7 + 25.4 = 92.1)
IPC Potential Based on Program Type Adjusted IPC Potential Based on Program Type2
High Cost Program
Possible Future Program with different ELCC
Current IPC Program
Possible Future Program
Pricing Program
Legend
Exhibit No. 1
Case No. IPC‐E‐21‐32
J. Ellsworth, IPC
Page 1 of 1