BPA Resource Program Draft Results May 10, 2018 B O N - - PowerPoint PPT Presentation

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

BPA Resource Program Draft Results

May 10, 2018

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Agenda

2

Draft - For Discussion Purposes Only

Time Mins Agenda Item 9:00 - 9:10 10 mins Introductions and Agenda Review 9:10 - 9:15 5 mins Project Overview 9:15 - 10:15 60 mins Recap of Model Inputs:

• Load Forecast
• CPA
• DR Potential Assessment

10:15 - 10:45 30 mins Needs Assessment Overview and Results 10:45 - 10:55 10 mins Break 10:55 - 11:15 20 mins Overview of Optimization Model 11:15 - 11:50 35 mins Draft Outputs and Results 11:30 - 11:45 10 mins Wrap Up and Next Steps

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

BPA Resource Program

• The BPA Resource Program
• Begins with a forecast of BPA load obligations and existing resources and then determines

needs

• Identifies and evaluates potential solutions to meeting those needs
• Energy efficiency, demand response, market purchases, wind, solar, gas plants, etc.
• Outlines potential strategies for meeting those needs
• The Resource Program is not
• A decision or policy document such as an Administrator’s Record of Decision
• A requirement of law or a regulating body such as FERC or NERC

3

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Resource Program Overview

4

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Previous Workshops

• Kick Off Meeting – March 2017
• End Use Load Forecast – June 2017
• CPA Kick off – May 2017
• CPA Methodology – September 2017
• Needs Assessment and Optimization Model Methodology – November 2017
• CPA Results Brownbag – March 2018

5

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

End Use Load Forecast

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Implementing End-Use Forecasting

• Goal: Implement End-Use forecasting to provide a frozen efficiency forecast for the BPA

Resource Program

• Process that is transparent to and involves our customers
• Consistency with the Council forecast data
• Representative of the BPA Power contractual obligation
• Integrates with our other BPA forecasting and planning activities
• Results: Contracting and Schedule complications required a small implementation in 2017
• Eleven customers had end-use (Statistically Adjusted End-use) models created due to

delay on ITRON product

• Remaining approximate 140 used existing time series process on pre-conservation

energy

Draft - For Discussion Purposes Only

7

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Forecast results

• Frozen Efficiency

forecast 2018-2040 annual growth rate averages 0.9%

• Approximately 600

aMW different in 2040

• Both will change as

we refine the forecasting method

• ver time

8

Draft - For Discussion Purposes Only

2000 4000 6000 8000 10000 12000 aMW

Frozen Efficiency and 2017 Expected Case

Frozen Efficiency Expected Case

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

SAE Methodology Benefits

• We learned and confirmed several things during this implementation process
• The process leverages work already done in the region using Council modeling

details while tailoring it to the BPA service territory

• The requisite review of the forecast is more thorough because of additional

assumptions to consider – Leads to a comprehensive review of economy activity by BPA staff – Review of additional assumptions when having local area discussions about the forecast – Augments thought process used in econometric approach

• Results are very utility specific

Draft - For Discussion Purposes Only

9

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Items to address over time

• Model Council forecast using this approach and see what results it produces to

make sure we calibrate correctly to Council details

• Determine method to provide link and details for future Conservation Potential

Assessments

• More work on customer specific reliable detail/data
• Create common vocabulary in region
• Arrange for details on distributed generation
• Work on data sources to leverage for or develop transmission level detail

Draft - For Discussion Purposes Only

10

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Needs Assessment

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Needs Assessment Overview

• The Needs Assessment provides forecasts of Federal system energy, capacity, and

balancing reserve needs

• Considers federal system resources and load obligations
• These results inform later steps of the Resource Program process
• The Needs Assessment methodology is largely unchanged from previous

Resource Programs

• Expanded to a 20 year continuous forecast
• Updated to a frozen efficiency load forecast

12

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Needs Assessment Metrics

• Annual Energy
• Evaluates the annual energy surplus/deficit under 1937-critical water conditions
• P10 Heavy Load Hour
• Evaluates the 10th percentile (P10) surplus/deficit over heavy load hours by month, given variability

in hydro generation, loads, and Columbia Generating Station output

• P10 Superpeak
• Evaluates the P10 surplus/deficit over the six peak load hours per weekday by month, given

variability in hydro generation, loads, and Columbia Generating Station output

• 18-Hour Capacity
• Evaluates the ability to meet the six peak load hours per day over three-day extreme weather

events assuming median water conditions

• Balancing Reserves
• Evaluates the ability to meet forecasted balancing reserve demand in the BPA balancing authority

area

13

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Results Summary – Energy

• Annual Energy deficits begin in fiscal year (FY) 2021 and grow to 850 aMW by 2039
• The largest P10 Heavy Load Hour deficits occur in winter, the first half of April, and fall
• January has the largest deficits (650 aMW in FY 2020 and 1,850 aMW in FY 2039)
• The largest P10 Superpeak deficits occur in winter, the first half of April, and late

summer/fall

• Near load-resource balance in FY 2020 and deficits grow to 1,000 aMW by FY 2039
• The P10 Superpeak deficits are smaller than the P10 Heavy Load Hour deficits in most

months, with the second halves of April and August being the exceptions

14

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Results Summary – Capacity & Reserves

• Winter is 18-Hour Capacity surplus over the study horizon, while summer is surplus in

FY 2020 and deficit thereafter (550 MW in FY 2039)

• The summer 18-Hour Capacity deficits are smaller than the P10 Superpeak deficits
• Demand for balancing reserves in the BPA balancing authority area is not expected to

reach 900 MW of incremental reserve over the study horizon

15

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Results: Annual Energy

• Key assumptions:
• Hydro generation based on 1937-critical water conditions
• Loads and Columbia Generating Station output reflect expected values

16

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Results: P10 Heavy Load Hour

• Key assumptions:
• Variability in hydro generation, loads, and Columbia Generating Station output

17

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Results: P10 Superpeak

• Key assumptions:
• Variability in hydro generation, loads, and Columbia Generating Station output

18

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Results: 18-Hour Capacity

• Key assumptions:
• Hydro generation based on median water

conditions

• Loads based on three-day extreme weather

event

• The 2015 Needs Assessment had surpluses of:
• 1,150 MW in winter
• 250 MW in summer

19

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Results: Balancing Reserves

• There is considerable uncertainty in future demand for balancing reserves, which includes:
• Amount and location of wind and solar development in the region
• Scheduling practices and elections
• Elections to self-supply
• New or expanded markets
• Departure of existing resources
• However, market conditions and RPS forecasts suggest future balancing reserve demand is

unlikely to reach 900 MW of incremental reserve

• Will continue to monitor the balancing reserves landscape and address the issue as

warranted in future Needs Assessments

20

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Needs Assessment Conclusion

• Overall, the results demonstrate that BPA is energy (fuel) limited
• P10 Heavy Load Hour deficits are larger than P10 Superpeak deficits in most months
• P10 Superpeak deficits are larger than 18-Hour Capacity deficits
• The most notable change from the 2015 Needs Assessment results is that the current

studies forecast summer 18-Hour Capacity deficits for most of the study horizon

21

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Conservation Potential Assessment

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Overview

• BPA has historically relied on the Council’s Power Plan for its assessment of

conservation potential in public power

• It is becoming increasingly important to have EE potential data specific to BPA
• In addition to Resource Program, the non-wires and Integrated Planning efforts
• BPA hired Cadmus to complete a BPA specific Conservation Potential Assessment
• Based on Council’s methodology
• Updated for changes since the passing of the 7th Plan (2016)

23

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Study Horizon

Seventh Plan: 2016-2035 Resource Program: 2020-2039 Conservation Potential Assessment

24 2016 2020 2035 2039

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Methodology

• Started with the 7th Plan baseline and measures
• Updated for:
• New codes and standards
• Expired measures
• Measures changes from RTF
• New Measures
• Savings accomplishments – 2016-2019
• Applied BPA specific characteristics (RBSA, CBSA, EIA)
• Estimated technical achievable potential

25

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Reminder

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Key Findings

1 2

• Look slightly

different from the region

3

• Identified a significant amount of

inexpensive energy efficiency

• The results were mostly

as expected

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Things We Learned about Public Power

01 02 03 04 05 06

Have more electric heating load Have 38% of all single family homes Have 36% of all commercial sq footage Have 48% of the industrial sales Have 34% of all irrigated acres

Have 30% of substations > 40,000 MWh

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Total Savings Potential

20 Year Cumulative Savings Potential Sector aMW % of Total Potential Agriculture 39 2% Commercial 542 30% Utility 67 4% Industrial 243 13% Residential 920 51% Total 1,812 100%

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Energy Efficiency Supply Curve

30

Draft - For Discussion Purposes Only

474 614 796 902 988 1,055 1,088 1,171 1,329 1,367 1,403 1,446 1,475 1,477 1,504 1,588 1,621 1,670 1,670 1,670 1,812 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000

Under $5 Under $10 Under $15 Under $20 Under $25 Under $30 Under $35 Under $40 Under $45 Under $50 Under $55 Under $60 Under $65 Under $70 Under $85 Under $100 Under $115 Under $130 Under $145 Under $160 Over $160

Cumulative 20-Year Potential - aMW

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Energy Efficiency - Inputs

Conservation potential provided to optimization model in 90 bundles

31

Draft - For Discussion Purposes Only

12 Levelized Cost Bundles Six End Uses Retrofit and Lost Opportunity

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Demand Response Potential Assessment

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

33

DR Potential Agenda

1. Overview and purpose 2. Basis for estimates 3. Estimate of DR Potential

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

34

• 1. Overview and Purpose
• This is the supply side (amount, cost) of DR over a 20 year horizon

with an expected seven year ramp

• This is not the needs or requirements side of the equation
• This is not a study of 2018 DR Potential or 2018 DR Costs

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

35

Overview and Purpose (Continued)

• This study was intended to fulfill the request for a DR Potential Assessment in

the 7th Power Plan

• Be a useful tool for BPA Resource Planning
• And it was expanded to include BPA Non-wires potential 2017 locational

considerations

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

36

Demand Response Potential Study

Geography (Primary):

• BPA public power portion of the region.
• I-5 Corridor/West of Cascades and East of

Cascades.

• Six geographical areas, as defined by

transmission planning.

Scope:

• All BPA preference customers.

Timing:

• 20-year potential estimates.
• Aligns with Resource Program and CPA.

Inputs:

• Interviews w/ 162 stakeholders (incl. 52 BPA

power customers) and 454 surveys.

• National and regional cost benchmarking.

Background: In the 7th Plan’s Action Plan, the Council recommended that the BPA conduct a study

• f DR potential and an assessment of barriers. The 2018 BPA Power Resource

Program will model DR as an input. The study helps answer these questions for BPA:

 How cost competitive is DR as an alternative for power (e.g. future peaks) and for transmission (e.g. build deferrals)?  How much (MWs) DR is there in BPA service territory and what will it take to get it?

BPA contracted and selected the Cadmus Group LLC (Cadmus) to perform this work. In addition to the Potential Assessment and Barriers Assessment, Cadmus performed a companion Elasticity Study .

DR Study Parameters

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

37

Scope of DR Analysis

BPA Preference Customer Peak Demand (2016) BPA Preference Customer Energy Sales (2016)

The analysis included all BPA Power preference customers, including federal agencies, direct-service industrial customers, tribal utilities, federal irrigation districts, and one port

• district. Figures above did not display 10 customers because their map boundaries were not available in BPA-sourced maps.

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

38

• 2. Basis for Estimates
• Considered East vs. West in Pacific Northwest
• Comparison of DR across USA and in Pacific Northwest
• DR costs are based on similar programs and achievable rates of

adoption

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

39

BPA Firm Power Customer Hourly Load Duration, East vs. West (2016)

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

40

DR Resources in Wholesale Markets of RTOs/ISOs and Peak Load Reduction

Note: None of these markets for DR are unconstrained; DR targets are pre-determined which limits the total amount of DR resources. Values are typical average within each area.

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

41

DR Capability as Percent of Peak Load – Public Utilities (2015)

Note: Few utilities reporting DR to EIA are unconstrained. Almost all of the DR reported is target-driven; reports indicate what utilities choose to do, not what is possible to accomplish.

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

42

• 3. Estimate of DR Potential
• Multiple DR products modeled using current technologies (and costs)
• Uses Ramp of 7 years, then overall 20 year levelized cost estimates*
• Sectors – Residential, Commercial, Industrial, Utility, Agricultural (DR

products have different costs)

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

43

Base Case Achievable Potential

The base case was developed by benchmarking research participation rates of common programs. These participation rates are generally a median value and are intended to depict participation in a healthy, established DR program. Most of the products reach a full ramp within 7 years, and after that grow with anticipated load rate changes. The base case values represent the mean of a range.

Area Winter Achievable Potential (MW) Percent of Area System Peak— Winter Summer Achievable Potential (MW) Percent of Area System Peak— Summer West 1,061 9.9% 807 10.8% East 490 9.6% 795 13.5% Total 1,551 9.8% 1,602 12.0%

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

44

DR Products Modeled

Sector DR Product Deployment Mechanism Seasonality Residential DLC—Water Heating DLC Summer and winter DLC—Space Heating DLC Winter only DLC—Central Air Conditioning (CAC) DLC Summer only DLC—Smart Thermostats DLC Summer and winter Critical Peak Pricing (CPP)* Tariff-Based Summer and winter Behavioral DR Direct Communication (e.g., event notifications) Summer and winter Commercial** DLC—CAC DLC Summer only Lighting Controls Automated Response Summer and winter Thermal Storage Cooling Storage Summer only Industrial*** Real Time Pricing (RTP)* Tariff-Based Summer and winter Commercial and Industrial Demand Curtailment and DLC Contract (Automated or Manual Response) Summer and winter Interruptible Tariff Tariff-Based Summer and winter Agricultural Irrigation DLC DLC Summer Utility System Demand Voltage Reduction (DVR) SCADA Summer and winter

*Cadmus assumed that Time of Use (TOU) rates were already in place. **In this assessment, Cadmus included public buildings in the commercial sector. ***In this assessment, Cadmus included public process loads such as municipal water treatment plants in the industrial sector.

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

45

Base Achievable DR – Product View

* The total achievable potential values in this detailed potential by product table do not match those in the previous slide because estimated achievable potential for DSI customer was estimated independently.

Product Winter Achievable Potential (MW) Percent of Area System Peak - Winter Levelized Cost ($/kW- year) Summer Achievable Potential (MW) Percent of Area System Peak - Summer Levelized Cost ($/kW- year)

Residential DLC—Space Heating 206 1.3% $53 0.0% n/a Residential DLC—Water Heating 389 2.5% $122 285 2.1% $167 Residential DLC—CAC 0.0% n/a 113 0.8% $74 Residential DLC—Smart Thermostat 222 1.4% $47 120 0.9% $88 Residential CPP 168 1.1% $10 57 0.4% $12 Residential Behavioral DR 37 0.2% $110 13 0.1% $111 Commercial DLC—CAC 0.0% n/a 110 0.8% $29 Commercial Lighting Controls 44 0.3% $32 55 0.4% $32 Commercial Thermal Storage 0.0% n/a 9 0.1% $51 C&I Demand Curtailment 184 1.2% $85 205 1.5% $85 C&I Interruptible Tariff 62 0.4% $73 69 0.5% $73 Industrial RTP 5 0.0% $35 5 0.0% $34 Agricultural Irrigation DLC 0.0% n/a 420 3.1% $44 Utility System DVR 225 1.4% $11 133 1.0% $12 Total 1,541 9.8% 1,592 11.9%

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

46

Winter Base Achievable Potential

Note: Displayed costs are the sum of annual costs, levelized over the 20-year study planning horizon, from the total resource cost perspective.

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

47

Summer Base Achievable Potential

Note: Displayed costs are the sum of annual costs, levelized over the 20-year study planning horizon, from the total resource cost perspective.

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Additional Resources

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

In Addition to EE and DR

The optimization also considered

• Market Reliance – How many purchases can be made from the market, where

the assumed market depth was determined by a supporting analysis and changes by HLH/LLH, month and year for the full duration of the study

• Solar – Fixed-axis and single-axis
• Wind – Assigned attributes representative of Columbia Basin wind resources
• Peaking Gas Plant – An LMS 100; GE’s popular aero-derivative gas plant

49

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Optimization Model

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

51

Draft - For Discussion Purposes Only

Optimization Model Overview

All resource options are evaluated against

• ur market price forecast on an hourly

basis for each risk iteration (Performance Run). The optimization then: 1) Solves for portfolio of resources that satisfy needs at the lowest cost (Portfolio 1). 2) Produces 39 additional portfolios by gradually increasing the budget constraint and minimizing total cost variation. This generates higher cost portfolios that include resources better aligned with BPA’s energy needs, lowering BPA’s exposure to the market.

Market Price Forecast Needs Assessment (HLH & Summer Capacity) Market Limits All Available Options Performance Run Optimization Portfolio 2 Portfolio 3 Portfolio 4 Portfolio 5 Portfolio 6 Portfolio 7 Portfolio 8 Portfolio 9 Portfolio 10 Portfolio …40 Portfolio 1 AURORA Process

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Additional Modeling Considerations

• Optimization results reflect assumptions that some resource options have negative costs
• Various benefits have been quantified and netted off total resource fixed costs, in some cases

benefits are sufficient to bring fixed costs below 0.

• Additionally, we are surplus energy on average and allow the model to sell all surplus energy at

the market price when its not needed to meet our needs. This expected revenue drives down total portfolio cost.

• Caveats:
• Portfolio values independent of rate impacts or actual expenditures required to acquire the

resource options

• Market prices are exogenous – they do not adjust as more purchases / sales are made
• Only one acquisition decision point modeled for all EE and DR programs (2020)
• We are not modeling the potential for new resources to alter hydro operations
• Did not include winter DR products because we do not have winter capacity needs

52

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Market Limits in AURORA

Given longer duration of this Resource Program and expected evolution of resource mix over the planning horizon, we have adopted a method that relies on AURORA. In order to ascertain market depth: 1. Start with our base resource build used to project future marginal costs of meeting load (market prices), this meets a 15% planning reserve margin in the PNW 2. Simulate scarcity conditions by reducing PNW hydro generation to monthly p10 level and allow all other risk models to operate normally (loads, transmission, wind, CGS, and natural gas prices) 3. Add incremental load increases to approximate greater resource retirements / fewer resource additions associated with higher levels of regional market reliance 4. On a monthly basis, determine level at which greater market reliance causes region to exceed 5% LOLP (as roughly approximated with AURORA) 5. Allocate a share of the market reliance to BPA and accept this as our market reliance limit

53

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Market Limits in AURORA

Start with baseline loads and resources This approach focuses on physical load-resource balance across the system, no modifications have been made to reflect frictions / improvements driven by changing market structures Incrementally increase PNW regional loads until loss-of-load events exceed threshold (5% LOLP) Determine BPA’s share (proportional to BPA load

• bligation / PNW regional

loads) and use this as our market reliance limit

54

Draft - For Discussion Purposes Only

BPA

PNW Loads PNW Loads PNW Loads

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

AURORA HLH Market Limits*

55

Draft - For Discussion Purposes Only

500 1000 1500 2000 2500 Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov 2020 2021 2022 2023 2024 2025 2030 2035

*In our analysis, no loss of load events occurred in LLH

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Outputs and Findings

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Resource Solutions - Efficiency Frontier

57

Draft - For Discussion Purposes Only

$100 $110 $120 $130 $140 $150 $160 $170 $180

Portfolio Total Cost Standard Deviation (Millions, NPV) Portfolio Total Cost (Millions, NPV)

Individual Portfolios

“Efficient” means that, at any given cost level, the indicated portfolio represents the one mix of resources which minimizes cost variance. Being on the Efficiency Frontier does not indicate equality amongst portfolios.

• Every portfolio meets our needs
• The model starts by determining the least cost portfolio
• $160M (over 20 years) is then added to each portfolio cost
• The model then reduces the variability in costs with different resource combinations given the new

cost levels

• Due to lower market exposure and electricity price levels, the reductions in cost variability are relatively

minor compared to the additional cost

Least Cost Least Cost Variability

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Efficiency Frontier

58

Draft - For Discussion Purposes Only

$100 $110 $120 $130 $140 $150 $160 $170 $180 Portfolio Total Cost Standard Deviation (Millions, NPV) Portfolio Total Cost (Millions, NPV)

Point where spending more money does not reduce variable cost exposure

Portfolios to the right could require a 6(c) process

1 2 3

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Resource Program Outputs – 2020/21

59

Draft - For Discussion Purposes Only

2020-2021 Totals

Portfolio Maximum Market Purchase (aMW) EE Aquired (aMW)* Highest EE Cost Bundle Demand Response (MW)

1 775 121 $25/MWh 40 2 737 154 $40/MWh 131 3 729 161 $50/MWh 131

Changes Across Portfolios

Change Total Market Purchase Total EE 2 year Total DR 2 year

1  2

• 38

33 91

2  3

• 8

7

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

60

Draft - For Discussion Purposes Only

Market Purchase Market Sales P10 HLH Energy Need

Monthly HLH Portfolio Energy vs P10 Need

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

• Anything else?

61

Draft - For Discussion Purposes Only

Monthly HLH Market Reliance vs Market Reliance Limit

Market Reliance Limit

Positive values indicate market purchases

Portfolio 1 Portfolio 2

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Demand Response Observations

62

Draft - For Discussion Purposes Only

Portfolio 2 Portfolio 1

MW

Notes:

• 40 MW by 2021
• 208 MW by 2025
• Is due its contribution to meeting

summer capacity needs in the least cost portfolio

• Portfolios 2-3
• Additional DR is added primarily

to reduce to cost variance

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

EE Portfolio Supply Curves

• Model takes all savings up to $20

in every portfolio

• To reduce cost variance, model

begins to move up EE supply curve

• Focuses on residential windows,

insulation and heat pumps in higher cost bins

63

Draft - For Discussion Purposes Only

20 40 60 80 100 120 140 160 180

$0/MWh and Under $0.01/MWh to $10/MWh $10.01/MWh to $20/MWh $20.01/MWh to $25/MWh $25.01/MWh to $30/MWh $30.01/MWh to $35/MWh $35.01/MWh to $40/MWh $40.01/MWh to $50/MWh $50.01/MWh to $60/MWh $60.01/MWh to $80/MWh $80.01/MWh to $100/MWh Over $100/MWh

aMW

Cumulative Levelized Cost by Portfolio

Port 1 Port 2 Port 3

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Portfolio 1 Acquisitions by End Use

64

Draft - For Discussion Purposes Only

200 400 600 800 1000 1200

$0/MWh and Under $0.01/MWh to $10/MWh $10.01/MWh to $20/MWh $20.01/MWh to $25/MWh $25.01/MWh to $30/MWh $30.01/MWh to $35/MWh $35.01/MWh to $40/MWh $40.01/MWh to $50/MWh $50.01/MWh to $60/MWh $60.01/MWh to $80/MWh $80.01/MWh to $100/MWh Over $100/MWh

Portfolio 1 Cumulative Savings - aMW

ELECTRONICS HVAC INDUSTRIAL LIGHTING OTHER WATER_HEATING

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Portfolio 2 Acquisitions by End Use

65

Draft - For Discussion Purposes Only

200 400 600 800 1000 1200

$0/MWh and Under $0.01/MWh to $10/MWh $10.01/MWh to $20/MWh $20.01/MWh to $25/MWh $25.01/MWh to $30/MWh $30.01/MWh to $35/MWh $35.01/MWh to $40/MWh $40.01/MWh to $50/MWh $50.01/MWh to $60/MWh $60.01/MWh to $80/MWh $80.01/MWh to $100/MWh Over $100/MWh

Portfolio 2 Cumulative Savings - aMW

ELECTRONICS HVAC INDUSTRIAL LIGHTING OTHER WATER_HEATING

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Which measures reduce cost variance?

• Green: industrial

energy management

• Red: residential

ductless heat pumps, windows and insulation

66

Draft - For Discussion Purposes Only

50 100 150 200 250

$0/MWh and Under $0.01/MWh to $10/MWh $10.01/MWh to $20/MWh $20.01/MWh to $25/MWh $25.01/MWh to $30/MWh $30.01/MWh to $35/MWh $35.01/MWh to $40/MWh $40.01/MWh to $50/MWh $50.01/MWh to $60/MWh $60.01/MWh to $80/MWh $80.01/MWh to $100/MWh Over $100/MWh

Cumulative Savings Additions – Portfolio 1 to 2 (aMW)

ELECTRONICS HVAC INDUSTRIAL LIGHTING OTHER WATER_HEATING

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

What We Learned

• BPA can continue to meet its needs with a combination of energy efficiency and

market purchases – no need for a major resource

• Demand Response appears to be an economically feasible option to summer

capacity needs

• All energy efficiency below market prices contribute to a least cost portfolio
• Not all savings are equal, measures shaped to meet Bonneville needs are

preferred over other savings

• The portfolio of savings selected differs from the current portfolio of savings

achievements

67

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Next Steps for EE

• BPA’s EE portfolio is a balance of many
• bjectives
• The results of the Resource Program

have provided Bonneville with new information to consider in program development

• Currently developing a proposal for how

to best address these objectives for the 2020/21 rate period

• June 12 Council Meeting
• June 21 IPR workshop

68

Draft - For Discussion Purposes Only

Equity BPA Needs Budget Emerging Technology Regional goals Customer Service

EE Portfolio

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Next Steps

• Public Comment Period
• May 11- June 1
• Finalize Resource Program
• July

69

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Appendix

70

Draft - For Discussion Purposes Only

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Appendix: P10 Heavy Load Hour (last 10 years)

• Key assumptions:
• Variability in hydro generation, loads, and Columbia Generating Station output

71

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Appendix: P10 Superpeak (last 10 years)

• Key assumptions:
• Variability in hydro generation, loads, and Columbia Generating Station output

72

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

73

DR Barriers Assessment

• While the DR Potential study provides a supply curve (amount, cost), the Barrier

Study considers barriers to adoption and mitigation strategies

• These two are Complementary Studies
• Barriers Study briefed earlier (key slide provided)
• Both based on strong research methods

B O N N E V I L L E P O W E R A D M I N I S T R A T I O N

Click to edit Master title style

74

Barriers to DR Development

Barrier Demand Response Distributed Generation Energy Storage SME n=17 STK n=12 PC n=25 DSP n=7 SME n=16 STK n=12 PC n=25 DSP SME n=16 STK n=12 PC n=24 DSP n=4 Economic/Market Lack of power customer business case 65% 75% 73% 86% 56% 83% 72% 81% 83% 76% 75% Lack of clearly defined need/value to BPA 59% 42% 64% 100% 56% 42% 56% 50% 50% 58% 75% Low power costs 56% 46% 70% 71% 59% 92% 85% 65% 58% 69% 25% Absence of organized market for DERs 61% 54% 59% 57% 13% 23% 24% 35% 46% 55% 50% Cost of development/ deployment 50% 46% 68% 29% 59% 77% 67% 88% 85% 89% 50% Lack of well-defined M&V framework 46% 18% 35% 14% 33% 27% 14% 50% 27% 41% 25% Organizational/Operational Competition for human/financial resources 63% 46% 58% 17% 43% 46% 39% 43% 36% 36% 25% Lack of staff knowledge and capability 44% 50% 30% 43% 47% 50% 19% 47% 58% 23% 0% Lack of standardized technical specs/agreements 35% 39% 48% 40% 20% 15% 29% 33% 25% 38% 0% Insufficient intra-organizational coordination/ communication 27% 50% 17% 29% 15% 40% 25% 23% 33% 22% 67% Infrastructure/Technology Data issues (e.g. lack of AMI, poor “big data” tools) 54% 39% 38% 60% 30% 25% 17% 50% 25% 30% 67% Back office systems 50% 60% 52% 0% 46% 30% 39% 46% 70% 38% 25% Communication protocols not standard; interoperability issues 36% 50% 48% 0% 18% 18% 17% 27% 46% 30% 25% Difficulty integrating DERs with current infrastructure 24% 23% 54% 20% 33% 31% 19% 47% 23% 36% 0% Concerns about cybersecurity 15% 20% 48% 14% 8% 20% 32% 8% 10% 33% 0% Lack of test facilities & infrastructure for communications to distributed devices 23% 27% 30% 0% 23% 18% 22% 31% 55% 18% 0% Ability to control/ manage EV charging and discharging 25% 33% 30% 20% 13% 11% 16% 14% 40% 30% 0% Unstable vendor supply chain 39% 25% 29% 20% 18% 17% 21% 46% 36% 39% 0% Legal/Regulatory Lack of established tariffs & contracts for DER 33% 63% 50% 60% 21% 44% 32% 39% 75% 35% 75% Concerns about data privacy 31% 27% 54% 14% 8% 9% 24% 8% 10% 29% 0% Environmental regulation/compliance and permitting/siting issues 0% 0% 24% 33% 42% 18% 0%

Source: Cadmus DER barriers rating survey Percent of respondents rating the barrier as a 4 or 5 on a 1 to 5 significance rating scale SME=BPA subject matter expert; STK=external stakeholder; PC= BPA power customer; DSP=DER service provider Note: Sample sizes identified are maximum sample size for each interview group and DER category. Due to small sample sizes, results should be interpreted as directional

Findings from 2017 Cadmus Barriers Assessment