system lcoe what are the costs of variable renewables
play

System LCOE: What are the costs of variable renewables? Falko - PowerPoint PPT Presentation

System LCOE: What are the costs of variable renewables? Falko Ueckerdt, Lion Hirth, Gunnar Luderer, Ottmar Edenhofer Working paper available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2200572 Paris, June 20, 2013 32th International


  1. System LCOE: What are the costs of variable renewables? Falko Ueckerdt, Lion Hirth, Gunnar Luderer, Ottmar Edenhofer Working paper available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2200572 Paris, June 20, 2013 32th International Energy Workshop falko.ueckerdt@pik-potsdam.de 1

  2. What are the costs of variable renewables? falko.ueckerdt@pik-potsdam.de 2

  3. Levelized costs of electricity (LCOE) are the full life-cycle costs (fixed and variable) of a technology per generation unit Projected LCOE for the US in cents per kWh Annual Energy Outlook 2013 (DOE/EIA) falko.ueckerdt@pik-potsdam.de 3

  4. Levelized costs of electricity (LCOE) are the full life-cycle costs (fixed and variable) of a technology per generation unit LCOE for plants commissioned in 2015 at 5% discount rate Projected Costs of Generating Electricity (IEA, 2010) falko.ueckerdt@pik-potsdam.de 4

  5. When will wind power be competitive? (investor)  „As soon as wind LCOE will be fallen below those of conventional plants!“ or equivalently  „When LCOE are equal to the average electricity price!“ What is the optimal amount of wind? (modeler/policy maker) LCOE of wind Average electricity price However, this is all wrong! falko.ueckerdt@pik-potsdam.de 6

  6. When will wind power be competitive? (investor)  „As soon as wind LCOE will be fallen below those of conventional plants!“  Don‘t compare LCOE of different technologies! or equivalently  „When LCOE are equal to the average electricity price!“  Don‘t compare generation costs with the average price! What is the optimal amount of wind? (modeler/policy maker) LCOE of wind Average electricity price However, this is all wrong! falko.ueckerdt@pik-potsdam.de 7

  7. Electricity is a unique commodity Demand is uncertain. Demand is variable, Ramping of plants is Transmission is costly storage is costly costly The value (price) of The value of electricity Short-term balancing electricity varies over varies between is costly time locations Electricity is a Heterogeneous w.r.t. Heterogeneous heterogeneous good lead-time between across space over time contract and delivery  Generation of different technologies is no perfect substitute, e.g. „nuclear power“ is not „wind power“  Do not compare LCOE falko.ueckerdt@pik-potsdam.de 8

  8. Variable renewables make electricity even more unique Bound to certain Output is fluctuating Output is uncertain locations • Wind speeds and • Winds and radiation • Resource quality solar radiation vary is uncertain day- varies geographically over time ahead • Thus its value • Thus its value • Forecast errors are depends on where it depends on when it costly is generated is produced „Profile costs“ „Balancing costs“ „Grid-related costs“ Integration costs (are neglected in LCOE comparison) falko.ueckerdt@pik-potsdam.de 9

  9. System LCOE are defined as the sum of generation and integration costs falko.ueckerdt@pik-potsdam.de 10

  10. Profile costs: Residual load duration curve shows challenges Load (GW) Low capacity credit Load duration curve Reduced full-load hours Residual load duration curve Hours of one year (sorted) Overproduction falko.ueckerdt@pik-potsdam.de 11

  11. Profile costs: Residual load duration curve shows challenges Wind Solar 100 100 Residual load (% of peak load) Residual load (% of peak load) 50 50 0 0 0% Wind 0% Solar PV 10% Wind 10% Solar PV 20% Wind 20% Solar PV -50 -50 30% Wind 30% Solar PV 40% Wind 40% Solar PV 50% Wind 50% Solar PV -100 -100 hours of a year (sorted) hours of a year (sorted) German data falko.ueckerdt@pik-potsdam.de 12

  12. System LCOE are defined as the sum of generation and integration costs Static Dynamic falko.ueckerdt@pik-potsdam.de 13

  13. When will wind power be competitive? (investor) What is the optimal amount of wind? (modeler/policy maker) System LCOE of wind LCOE of wind Average price falko.ueckerdt@pik-potsdam.de 14

  14. When will wind power be competitive? (investor) What is the optimal amount of wind? (modeler/policy maker) System LCOE of wind = marginal cos costs of of „powe wer“ when produced from wind plant LCOE of wind = margin inal costs ts o of „win ind“ Average price = margin inal l value o e of „power er“ Interpretation • There are two goods “wind” and “power” • „Power“ is perfectly matching load Additional costs needed to transform „wind“ into „power“  integration costs • • The total costs to cover load with „wind“ are System LCOE falko.ueckerdt@pik-potsdam.de 15

  15. Two perspectives, one framework: System LCOE and market value Market Value Perspective System LCOE Perspective €/MWh €/MWh Integration costs Average Profile Balancing Grid- Wind Wind’s Profile Balancing Grid- Wind’s electricity Costs Costs related Market LCOE Costs Costs related System price Costs Value Costs LCOE marg rginal v l value marg rginal v l value marg rginal c l costs marg rginal l costs s of „ „powe wer“ of „ „wi wind“ of „ „wi wind“ of „ „powe wer“ falko.ueckerdt@pik-potsdam.de 16

  16. Profile Costs: The “Cannibalization” Effect falko.ueckerdt@pik-potsdam.de 17

  17. Market value of wind decreases with increasing share Integration costs increase falko.ueckerdt@pik-potsdam.de 18

  18. Two perspectives, one framework: System LCOE and market value System LCOE of wind = marginal cos costs of of „powe wer“ when produced from wind plant LCOE of wind = margin inal costs ts o of „win ind“ Integration costs Average price = margin inal l value o e of „power er“ Market value = marg rginal l value o e of „wi wind“ falko.ueckerdt@pik-potsdam.de 19

  19. System LCOE – magnitude and shape 140 • From literature: Grid and balancing costs Wind: System LCOE (Holttinen et al. 2011; Gross et al. 2006; 120 Hirth 2012a, dena 2010) 100 • From a simple model: profile costs. Integration • Parameterized from German data, costs 80 representative for thermal systems in Europe 60 System LCOE • Caveats that increase integration costs Short-term System LCOE 40 LCOE Generation • No import/export Long-term capacity adjustment costs Grid costs 20 • No demand elasticity Balancing costs Profile costs Generation costs • No storage 0 5 10 15 20 25 30 35 40 • Power sector only Final electricity share of Wind (%) Final elecitricity share of wind (%) • Integration costs of wind power can be in the same range as generation costs at moderate shares (~20%) • A significant driver of integration costs are profile costs, especially the reduced utilization of capital-intensive thermal plants.  Integration costs can become an economic barrier to deploying VRE at high shares.  An economic evaluation of wind and solar power must not neglect integration costs. falko.ueckerdt@pik-potsdam.de 20

  20. The market value (here value factor) reduces: Market Data, Model Results, Literature Review 1. 2. 3. At 30% penetration, the value factor of wind falls to 0.5 – 0.8 of the base price. In Germany, it has already fallen from 1.02 to 0.89 as penetration increased from 2% to 8%. Hirth, Lion (2013): "The Market Value of Variable Renewables", Energy Economics 38, 218-236. falko.ueckerdt@pik-potsdam.de 21

  21. Conclusions 1. We show that LCOE are a flawed indicator to evaluate power technologies 2. We propose a better metric ‘System LCOE’ 3. We present a framework of System LCOE and market value and link it to welfare theory 4. Integration costs of wind power can be in the same range as generation costs at moderate shares (~20%) 5. A significant driver of integration costs are profile costs, especially the reduced utilization of capital-intensive thermal plants. 6. Integration costs can become an economic barrier to deploying VRE at high shares. 7. An economic evaluation of wind and solar power must not neglect integration costs. falko.ueckerdt@pik-potsdam.de 22

Recommend


More recommend