1. CVP Cost Allocation Public Workshop – January 18, 2013 “PLEXOS Methodology and Assumptions”

2. Methodology Summary Estimate value of CVP power by comparing the differential costs for two scenarios: With fully-functional CVP Without CVP, but with replacement portfolio Study is performed with CVP constraints modeled PLEXOS is used to determine the difference in variable costs between the two scenarios Capital and fixed operating costs are from another source

3. PLEXOS Overview Fundamental market simulation model (supply and demand) Minimizes total market cost for all variables: Energy and ancillary services (AS) Fuel and variable operating expenses Emission costs (if modeled) Wheeling costs and losses Subject to 1000s of constraints: System load and AS Plant performance Transmission capability Uncertainty of variable-energy resources

4. Solving UC/ED using MIP Unit Commitment and Economical Dispatch can be formulated as a linear problem (after linearization) with integer variables of generator on-line status Minimize Cost = generator fuel and VOM cost + generator start cost + contract purchase cost – contract sale saving + transmission wheeling + energy / AS / fuel / capacity market purchase cost – energy / AS / fuel / capacity market sale revenue Subject to –Energy balance constraints –Operation reserve constraints –Generator and contract chronological constraints: ramp, min up/down, min capacity, etc. –Generator and contract energy limits: hourly / daily / weekly / … –Transmission limits –Fuel limits: pipeline, daily / weekly/ … –Emission limits: daily / weekly / … –Others

5. Integration of Mid- and Short-Term Constraints PLEXOS includes three integrated algorithms: –Long-, mid-, and short-term –Three perspectives are seamlessly integrated –Mid-term simulation decomposes hydro, fuel, emission, and energy constraints for the short-term simulation –CALSIM monthly output decomposed into daily amounts for short-term Long-term security assessment Maintenance planning and outage assessment Mid-term simulation Resolve and price emission /fuel/ energy constraints Short–term simulation Full chronological simulation

6. Detailed Generator Modeling General chronological constraints modeled, i.e., –Minimum up and down time –Ramp up and down rate –Minimum capacity with hourly economic or must-run status –Reserve (regulation up/down, spinning and non-spinning) provision capacities –Start cost as a function of number of hours being down –Forbidden operation zone User-specified fuel mixture / mixture ranges or model-determined fuel mixture Heat Rate as a function of fuel types –Average heat rate for multiple loading points –Incremental heat rate for multiple loading points –Polynomial fuel-generation IO curve Emission rate with removal rate Initial commitment and dispatch status

7. Combined Cycle Modeling, continued HR=10316 Btu/kWh Efficiency=33% Boiler efficiency = 80% ~ Gen=160 MWh Fuel=1.68e+9 Btu Duct Burner Fuel=1.45e+8 Btu HR=10500 Btu/kWh Efficiency=32.5% HR=10500 Btu/kWh Efficiency=32.5% Energy content of electricity = 3412 Btu/kWh Waste=1.134e+9 Btu 1.96004e+9 Btu ~ ~

8. PLEXOS Hydro Modeling Inflows, storages, plants, and spills are modeled and optimized on an hourly basis –In terms of either acre-feet (volume) or MWh The hydro contribution is maximized given energy and AS markets (or system requirements) Hydro is fully integrated with the thermal (hydro- thermal integration) and is perfectly arbitraged against all available markets

9. An Example of Cascaded Hydro System Inflow Storage II ~ P/S 2 Storage III Storage V Storage I Sea Inflow ~ P/S 1 ~ P/S 3 ~ H 2 ~ H 1 ~ H 3 ~ H 6 ~ H 4 ~ H 5 Storage VI

10. LT-Plan: PLEXOS for Integrated Resource Planning Alternative portfolio development methodology Objective: Minimize net present value of forward-looking costs (i.e. capital, fixed operating and production costs) Production Cost P(x) Capital Cost C(x) Total Cost C(x) + P(x) Cost ($) Investment xOptimal Investment x*

11. Hydro Value Energy Ancillary services Fast ramp (up and down) No greenhouse gas

12. Primary Data Sources WECC TEPPC (Transmission Expansion Planning Policy Committee) regional database (version PCO) CA Utility LTPP (Long-Term Power Plan) revisions and updates for CA CVP-specific information

13. Selected Examples of Data Input to PLEXOS Simulation year – 2020 Base year for dollars – 2010 CA hydro aggregated in two zones –Northern and Southern California –CVP extracted from aggregated hydro

14. Data Inputs II CAISO 2011-2020 Changes (MW) –Summer Peak Load Summer peak load = 6,200 MW Demand-side reductions = 8,100 MW Net peak summer load = (1,900 MW) – Summer Generation Capacity Retirements = 13,100 MW New additions = 11,100 MW (Thermal, RPS) Net summer capacity = 2,000 MW –primarily RPS and OTC replacement

15. Data Inputs III CAISO 2011-2020 Changes (MW) –Renewable Portfolios In-state = 14,200 MW Out-of-state = 5,093 In-state renewable types – Hydro = 0 MW – PV Solar = 4,600 MW – Solar Thermal = 3,600 MW – Wind = 5,034 MW Out-of-state renewable Wind = 5,000 MW

16. Data Inputs IV Natural gas prices (2010 $) –PG&E Citygate -- $5.61 / MMBtu (delivery to burner-tip adds 7 to 23 cents / MMBtu) –SoCal Border -- $5.41 / MMBtu (delivery adds 44 cents / MMBTU) –Current Price (1/4/2013) -- $3.30 to $3.60 / MMBtu (source: California Energy Markets) CO2 Emission Price (2010 $) –$36.30 (short-ton CO2) CA Net Exchange (summer peak) –16,400 MW

17. Questions ?

18. Selected Acronyms AS – Ancillary Services CAES – Compressed air energy storage CAISO – CA Independent System Operator CO2 – Carbon dioxide LTPP – Long-Term Procurement Plan OTC – Once-Through Cooling TEPPC – Transmission Expansion Planning Policy Committee (WECC regional database for market simulation purposes) WECC – Western Electric Coordinating Council

19. References LTTP data assumptions – http://www.caiso.com/Documents/2011-08- 10_ErrataLTPPTestimony_R10-05-006.pdf