NorthWestern Energy Group, Inc. · NASDAQ:NWE

NorthWestern Energy — Technical Presentation (January 2016)

84 slides from NorthWestern Energy's technical presentation, Jan 25, 2016

1. Slide 1: 2015 Electric Supply Resource Procurement Plan – ETAC Meeting January 25, 2016

   2015 Electric Supply Resource Procurement Plan – ETAC Meeting January 25, 2016

2. Slide 2: Assumptions Update

   Assumptions Update • Assumption Updates & Introduction 9:00-9:20 • Distinguishing Factors of 2015 Resource Plan 9:20-9:30 • Integrated System 9:30-10:20  Integrating the Hydro’s and thermal generation  Regulation requirements • Near-term Resource Additions 10:20-10:30 • Avoided Costs 10:45-10:55 •

3. Slide 3: Assumptions Update

   Assumptions Update 3 ascend analytics

4. Slide 4: Generic Resource Costs

   Generic Resource Costs Changes since last ETAC meeting: $/kW-yr Cause of Change ICE -100 • Consistent basis of cost determination • Consistent delta in ICE and CT costs Frame CT +100 • Change in exchange rate 4 ascend analytics

5. Slide 5: Commodity Prices 2015 Plan

   Commodity Prices 2015 Plan • Natural Gas and Electricity Prices  Updated market forward prices of natural gas and electric prices: • Originally prices marked July 6, 2015 • Updated prices marked December 9, 2015  Extension allowed NorthWestern to update foreword market prices  More accurately ref

6. Slide 6: Commodity Escalation

   Commodity Escalation • Commodities that will be escalated with the EIA escalation rate  Natural gas  Electricity – maintaining structural relationship between natural gas and electricity  Carbon • Commodity that won’t be escalated with the EIA escalation rate  Coal  Current Colstrip operating p

7. Slide 7: Distinguishing Factors

   Distinguishing Factors 7 ascend analytics

8. Slide 8: Distinguishing Factors 2015 Electric Resource Plan

   Distinguishing Factors 2015 Electric Resource Plan • Integrated System Operations  Co-optimize Hydro’s and DGGS for regulation  Joint optimization of all resources to meet load following ramps • Resource Adequacy  Evaluate NorthWestern relative to traditional planning criteria  Evaluation of cap

9. Slide 9: Integrated System

   Integrated System 9 ascend analytics

10. Slide 10: Current Sources of Regulation and Load Following

    Current Sources of Regulation and Load Following • DGGS provides all of NorthWestern’s regulation and load following requirements 10 ascend analytics

11. Slide 11: Integrated System for Regulation Supply

    Integrated System for Regulation Supply • Integrating the Hydros to provide regulation enhances system capabilities  Lowers costs  Provides greater flexibility necessary to manage renewable resources 11 ascend analytics

12. Slide 12: Integrated System for Load Following

    Integrated System for Load Following • Integrating the Hydros to provide regulation enhances system capabilities  Lowers costs  Provides greater flexibility necessary to manage renewable resources 12 ascend analytics

13. Slide 13: Regulation Service Definition

    Regulation Service Definition • Regulation maintains interconnection frequency, manage differences between actual and scheduled power flows, and match generation to load. • Regulation has the property of netting to approximately zero over the course of several hours. Regulation tracks rapid movement

14. Slide 14: Ancillary Service Definitions

    Ancillary Service Definitions Definition of Regulating and Contingency Reserves Service Description Response Speed Duration Cycle Time Normal Conditions Regulating Reserve Online resources, on automatic generation control, that can respond rapidly to system-operator requests for up and down movement

15. Slide 15: Area Control Error for Non-ISO Evaluations

    Area Control Error for Non-ISO Evaluations • Frequency does not change in an interconnection when there is a balance between resources and customer demand Example: Balancing Authority (BA) with Bias = -50 MW/0.1Hz buys 300 MW. Actual flow into BA =310 MW. Frequency is 60.01 Hz. ACE=(-310- -300)-10*(

16. Slide 16: Flexibility Considerations

    Flexibility Considerations • The effect of renewable integration on flexibility needs  Delineate regulation from load following requirements  Estimate system flexibility requirements to integrate renewables • Cost of renewable integration  Generation costs of regulating resources  Flexible resou

17. Slide 17: The Effect of Current Renewables on Regulation Requirements

    The Effect of Current Renewables on Regulation Requirements Majority of regulation required for renewable integration Target Regulation ≈ 33 MW Date and Hour 17 ascend analytics

18. Slide 18: 1-Hour Ramps

    1-Hour Ramps 18 ascend analytics

19. Slide 19: The Effect of Doubled Renewables on Regulation Requirements

    The Effect of Doubled Renewables on Regulation Requirements Regulation requirements increase substantially with added renewables Target Regulation ≈ 62 MW 19 ascend analytics

20. Slide 20: 1-Hour Ramps Under Doubled Renewables

    1-Hour Ramps Under Doubled Renewables 20 ascend analytics

21. Slide 21: The Effect of Wind on Regulation

    The Effect of Wind on Regulation 21 ascend analytics

22. Slide 22: Regulation Cost

    Regulation Cost 22 ascend analytics

23. Slide 23: Renewable Integration Cost DGGS Provides Regulation

    Renewable Integration Cost DGGS Provides Regulation 23 ascend analytics

24. Slide 24: Sources of Ancillary Services 50/50

    Sources of Ancillary Services 50/50 24 ascend analytics

25. Slide 25: Minutely Regulation Chart

    Minutely Regulation Chart 25 ascend analytics

26. Slide 26: Resource Adequacy

    Resource Adequacy 26 ascend analytics

27. Slide 27: Montana Planning Criteria

    Montana Planning Criteria 69-8-419. Electricity supply resource planning and procurement -- duties of public utility -- objectives -- commission rules. 1. The public utility shall: a) Plan for future electricity supply resource needs; b) Manage a portfolio of electricity supply resources; and c) Pro

28. Slide 28: Existing Generation Capacity vs. Peak Load, Load + 15% Reserve Margin

    Existing Generation Capacity vs. Peak Load, Load + 15% Reserve Margin 28 ascend analytics

29. Slide 29: Projected Loss of Load Under Current Portfolio (MW and Hours)

    Projected Loss of Load Under Current Portfolio (MW and Hours) 29 ascend analytics

30. Slide 30: Summary

    Summary NorthWestern Energy 30 | ascend analytics

31. Slide 31: Addressing Resource Needs

    Addressing Resource Needs • Adding Flexibility—more cost effective integration of renewables  Lower cost delivery of regulation  Enhanced ramping capabilities • Progressing incrementally toward resource adequacy  Incremental steps to help ensure future reliability  Reduce future supply risks and

32. Slide 32: Systematically Incorporating Uncertainty

    Systematically Incorporating Uncertainty • Heavy dependence on deterministic results with scenarios  The likelihood of result is not understood  Model inputs are variable and interdependent  Deterministic modeling misses critical scenarios producing an inconsistent value • What’s the impact of un

33. Slide 33: NPV of Portfolio Costs Under Near-Term Resource Options and Optimal Expansion Plan

    NPV of Portfolio Costs Under Near-Term Resource Options and Optimal Expansion Plan 33 ascend analytics

34. Slide 34: Risk premium monetizes value of future risks

    Risk premium monetizes value of future risks • The risk premium is the probability-weighted average of costs exceeding the mean  The risk premium monetizes the value of risk into a discrete cost  Applies same analytic techniques used in insurance industry and for real option pricing • Levels the p

35. Slide 35: Optimal Expansion Planning

    Optimal Expansion Planning 35 ascend analytics

36. Slide 36: Deterministic Capacity Expansion Optimization: Best Athlete

    Deterministic Capacity Expansion Optimization: Best Athlete The best expansion plan for each scenario is akin to the best athlete for each sport. Which is the best plan for overall scenarios? Different conditions yield Michael Phelps Matt Biondi Ryan Lochte Gary Hall, Jr. different optimal plans Obj

37. Slide 37: Stochastic Capacity Expansion Optimization: Best Triathlete

    Stochastic Capacity Expansion Optimization: Best Triathlete Best Expansion Plan for All Future Simulated States Requirements • Robust simulations • Advanced optimization of energy and ancillary services Advantages • Simplifies decision choices • Captures full tail of cost distribution • Accounts for

38. Slide 38: Optimal Capacity Expansion Plan

    Optimal Capacity Expansion Plan Min NPV of Rev Requirements while satisfying increasing reserve margin constraint 38 ascend analytics

39. Slide 39: Optimal Expansion Plan: 50% Load Growth

    Optimal Expansion Plan: 50% Load Growth Min NPV of Rev Requirements while satisfying increasing reserve margin constraint 39 ascend analytics

40. Slide 40: Optimal Expansion Plan: 150% Load Growth

    Optimal Expansion Plan: 150% Load Growth Min NPV of Rev Requirements while satisfying increasing reserve margin constraint 40 ascend analytics

41. Slide 41: Effect of Retiring Colstrip on Capacity

    Effect of Retiring Colstrip on Capacity 41 ascend analytics

42. Slide 42: Comparison of Expansion Plans in Different Future Scenarios

    Comparison of Expansion Plans in Different Future Scenarios 42 ascend analytics

43. Slide 43: NPV of Expansion Plan Portfolio Costs Under Different Load Growth Scenarios

    NPV of Expansion Plan Portfolio Costs Under Different Load Growth Scenarios * * No decommissioning costs included. 43 ascend analytics

44. Slide 44: NPV of Expansion Plan Portfolio Costs Under Different Price Scenarios

    NPV of Expansion Plan Portfolio Costs Under Different Price Scenarios 44 ascend analytics

45. Slide 45: "What if" Expansion Plan Scenarios

    “What if” Expansion Plan Scenarios 45 ascend analytics

46. Slide 46: NPV of Portfolio Costs Under Near-Term Resource Options and Optimal Expansion Plan

    NPV of Portfolio Costs Under Near-Term Resource Options and Optimal Expansion Plan 46 ascend analytics

47. Slide 47: Integrated System

    Integrated System 47 ascend analytics

48. Slide 48: Value of Proposed Operating Changes 2015 Electric Resource Plan

    Value of Proposed Operating Changes 2015 Electric Resource Plan • Resource Adequacy  Substantial capacity short-fall to meet load  Current system lacks flexibility to efficiently manage future renewable resources • Integrated System Operations  Joint optimization of all assets increases system fl

49. Slide 49: Sim Engine Validation

    Sim Engine Validation 49 ascend analytics

50. Slide 50: Integrating Physical and Financial Uncertainty Hybrid Modeling Framework

    Integrating Physical and Financial Uncertainty Hybrid Modeling Framework Meaningful Uncertainty • Produce realistic simulations of physical system and market conditions • Validation of simulations • Simulate physical variables that drive market prices During delivery simulations Wx Sim Renewables Op

51. Slide 51: Confidence Intervals and Simulation Spread

    Confidence Intervals and Simulation Spread Forward Price Simulations Mid-C Heavy price simulations spread from the mean • Simulated forward prices are plotted over time for the mean, 5th, and 95th percentiles • Expect that uncertainty will grow with time • F0=E(Ft)=E(FT)=E(St) Confidence Intervals f

52. Slide 52: Fuel Price Simulations

    Fuel Price Simulations • Simulated forward prices are Colstrip Fuel Price plotted over time for the mean, $1.30 $1.20 5th, and 95th percentiles $1.10 $/MMBtu • Seasonal changes in price and $1.00 $0.90 uncertainty $0.80 $0.70 $0.60 1/1/2016 2/1/2017 3/1/2018 4/1/2019 5/1/2020 6/1/2021 7/1/2022 8/1/2

53. Slide 53: Forward Price Validation

    Forward Price Validation Risk Bands for Implied On-Peak Heat Rates Maintaining Correlations • Simulations maintain implied heat rate distributions • Correctly value generation assets • Preserve proper value of portfolio risks • Enables proper value of cross hedges 120 12 100 10 80 8 60 6 Simulated f

54. Slide 54: CO2 Price Distribution

    CO2 Price Distribution • In this analysis, CO2 prices are modeled beginning in year 2022 • CO2 prices in $/ton are modeled using a triangular distribution with a minimum of 0, a mode of 20, and a maximum of 40 in 2022 • These CO2 price assumptions are similar to the 2013 Plan starting a year later C

55. Slide 55: Weather Hydro Flows Load Price Simulations

    WeatherHydro FlowsLoadPrice Simulations Hydro Flows Gas Electric Weather Load Price 55 ascend analytics

56. Slide 56: Weather-Load Relationship

    Weather-Load Relationship Weather-Load Validation Maintaining Correlations Simulated vs. Historical • Incorporating weather into the load model maintains integrity in the weather – load relationship • Simulations nicely smooth out “bumps” of historical weather record • Simulations provide for new ex

57. Slide 57: Load – Evaluating Difference 5th, Mean, 95th

    Load – Evaluating Difference 5th, Mean, 95th Load Simulation Validation Results • Demonstrate a robust model to simulate future load • Capture uncertainty in load growth with extreme weather scenarios validated prior to a load simulation • Validated monthly load values • Validated hourly load values

58. Slide 58: Load Validation

    Load Validation August Load Confidence Intervals • Confidence intervals for hourly load by month at the mean, 5th, and 95th percentiles • Alignment of simulations with historical data January Daily Load Shape • Morning and evening peak during cold months • Single afternoon peak during warm months 58

59. Slide 59: Avoided Costs

    Avoided Costs 59 ascend analytics

60. Slide 60: Avoided Cost

    Avoided Cost • Case 1: Selling Power  Generation greater than load  Avoided cost = marginal cost of generation to serve load • Case 2: Buying Power Short economic energy  Avoided cost = market price of power • Case 3: Must Run Generation > Load  Avoided cost = nothing, because unneeded power, m

61. Slide 61: Solving for Capacity Expansion Plans

    Solving for Capacity Expansion Plans 61 ascend analytics

62. Slide 62: Solving for Capacity Expansion Plans: Objective Functions

    Solving for Capacity Expansion Plans: Objective Functions Min Expected Value of Total Cost 𝑇𝑇 𝑁𝑁 E[NPV of Total Costs] = E � � 𝛽𝛽𝑡𝑡 ∗ 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 ∗ 1𝑖𝑖𝑖𝑖 𝑡𝑡=1 𝑖𝑖=1 • Fixed Costs follow revenue requirements  Depreciation, Amortization, Current…

63. Slide 63: Solved using dynamic programming

    Solved using dynamic programming • Example: portfolio can build CC, wind, or both  Load increases in t = 3  RPS takes force in t = 5 Represents a portfolio configuration (‘state’) • State space including infeasible states shown here End t=1 t=2 t=3 t=4 t=5 Nothing built Gas, no wind Wind, no gas B

64. Slide 64: Solved using dynamic programming

    Solved using dynamic programming Identify and remove infeasible states Infeasible because of need to meet increased load starting from t = 3 End t=1 t=2 t=3 t=4 t=5 Nothing built Gas, no wind Wind, no gas Infeasible because of need to meet RPS starting from t = 5 Both wind and gas 64 ascend analytic

65. Slide 65: Enumerate costs for states

    Enumerate costs for states Gross margin shown in green t=1 t=2 t=3 t=4 t=5 End 0 0 Positive profit for CC Nothing out 100 100 150 Gas CC 0 0 30 50 LM6000 0 0 0 25 CT Frame F 65 ascend analytics

66. Slide 66: Enumerate fixed costs for transitions between states

    Enumerate fixed costs for transitions between states Fixed costs sum linearly  Building a CC is 200  Wind contract has up-front cost of 50 Allowable transition (e.g., build an asset) – ‘feasible action’ t=1 t=2 t=3 t=4 t=5 End 0 0 0 Transition fixed costs from only a few Nothing built states shown

67. Slide 67: Backward pass

    Backward pass Start from next-to-last time period (t = 4) Begin by enumerating feasible action set for each state t=1 t=2 t=3 t=4 t=5 End 0 0 Two feasible actions from Nothing built this state: build CC, or One feasible action from do not build CC. this state: build wind. (Need to meet RPS, and -100

68. Slide 68: Backward pass

    Backward pass Calculate cost associated with each feasible action  Quantity includes both fixed cost of this action, and future costs t=1 t=2 t=3 t=4 t=5 End 0 0 Nothing built -100 -100 -100 Gas, no wind 100 150 150 150 150 Wind, no gas 250 50 50 50 Both wind and gas 50 68 ascend analytics

69. Slide 69: Backward pass

    Backward pass Select lowest-cost action from each state in t = 4  Retain lowest cost action, and discard suboptimal actions t=1 t=2 t=3 t=4 t=5 End 0 0 Nothing built -100 -100 -100 Gas, no wind 100 150 150 150 150 Wind, no gas 50 50 50 Both wind and gas 50 69 ascend analytics

70. Slide 70: Backward pass

    Backward pass The reverse partial objective function sum is calculated for each of the states in the t = 4 time period. t=1 t=2 t=3 t=4 t=5 End 0 0 The reverse partial cumulative sum at a state in time represents the cumulative cost going forward Nothing built from that state in time, for optimal de

71. Slide 71: Backward pass

    Backward pass The process is repeated, starting from the previous time period. t=1 t=2 t=3 t=4 t=5 End 0 0 Nothing built -100 -100 0 0 Gas, no wind 50 150 150 300 0 Wind, no gas 200 50 50 100 Both wind and gas 0 71 ascend analytics

72. Slide 72: Backward pass

    Backward pass The process is repeated, starting from the previous time period. t=1 t=2 t=3 t=4 t=5 End 0 0 Nothing built -100 -100 0 Gas, no wind 150 150 150 300 Wind, no gas 300 50 50 Both wind and gas 100 72 ascend analytics

73. Slide 73: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 0 If there is a tie, either branch can be taken; Nothing built one branch is shown for visual clarity. -100 -100 Gas, no wind 150 450 Wind, no gas 50 150 Both wind and gas 73 ascend

74. Slide 74: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 0 Nothing built 200 -100 -100 0 Gas, no wind 50 50 150 450 0 Wind, no gas 200 250 50 150 Both wind and gas 0 74 ascend analytics

75. Slide 75: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 0 Nothing built 100 -100 -100 Gas, no wind 200 500 150 450 Wind, no gas 350 400 50 Both wind and gas 150 75 ascend analytics

76. Slide 76: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 0 Nothing built 100 -100 -100 Gas, no wind 150 Wind, no gas 350 50 Both wind and gas 150 76 ascend analytics

77. Slide 77: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 100 Nothing built -200 Gas, no wind 500 Wind, no gas 200 Both wind and gas 77 ascend analytics

78. Slide 78: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 100 0 Nothing built 200 50 -200 Gas, no wind 500 Wind, no gas 250 200 Both wind and gas 78 ascend analytics

79. Slide 79: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 100 Nothing built 0 550 Gas, no wind Wind, no gas 450 Both wind and gas 79 ascend analytics

80. Slide 80: Backward pass

    Backward pass All time periods are iterated through reverse sequentially (backward in time). t=1 t=2 t=3 t=4 t=5 End 0 Nothing built 0 Gas, no wind Wind, no gas Both wind and gas 80 ascend analytics

81. Slide 81: Backward pass completed

    Backward pass completed t=1 t=2 t=3 t=4 t=5 End 0 Nothing built Gas, no wind Wind, no gas Both wind and gas 81 ascend analytics

82. Slide 82: Forward pass

    Forward pass After the backward pass is completed, the optimal path is found by iterating forward through time from the starting state. Starting state t=1 t=2 t=3 t=4 t=5 End Nothing built Build CC Do nothing Do nothing Gas, no wind Build wind Wind, no gas u* Both wind and gas The collection of acti

83. Slide 83: Scenarios Considered

    Scenarios Considered Study Portfolio 50/50 Base Case Base portfolio, no new units. Optimal Expansion Plan DGGS reg Selected CapEx Res Margin Escalating Limits col mustrun wind Optimal Expansion Plan Wartsila Reg Replace 2 wartsilas with "reg wartsilas". Replace DGGS with DGGS econ dispatch. Frame CT

84. Slide 84: Final Thoughts

    Final Thoughts • Additional Discussion • Next ETAC Meeting 84 ascend analytics

NorthWestern Energy Group — Technical Presentation (January 2016) on 6ix