16_Economic_assessment_of_PV_and_wind_for_energy_planning_Cairo_Egypt.pdf
0 likes33 views
The document discusses levelized cost of electricity (LCOE) analysis for photovoltaic and wind energy projects. It provides worked examples of how LCOE can be used to assess the sensitivity of project costs to changes in key input parameters. The examples show that LCOE is highly sensitive to wind speed, solar irradiation data quality, and other economic assumptions. Relying on lower quality meteorological data can significantly increase estimated LCOE and power purchase agreement costs compared to using high-quality ground measurement data. Accurate resource and cost data are therefore critical for effective energy policymaking and project development.
16_Economic_assessment_of_PV_and_wind_for_energy_planning_Cairo_Egypt.pdf
- 1. Session 3: Economic assessment of PV and wind for energy planning IRENA Global Atlas Spatial planning techniques 2-day seminar
- 2. Central questions we want to answer 1. Once we know how much electricity can be produced in our country with given resources (technical potential), we will be able to estimate their generation costs 2. As all available data comes with uncertainties, we should know a. how sensitive results react on changing input parameters, and, b. what socio-economic effect highly uncertain input data could have. 2
- 3. 3 © RENAC 2014 Irradiation on tilted plane (Wh/m²/a) Energy generation costs at specific site (€/Wh) Conversion horizontal solar radiation to optimally tilted plane Optimal tilt angle Energy output calculation Pre-conversion losses Conversion losses System losses (%) CAPEX OPEX WACC Life time Economic parameters (PV plant and grid connection) Annual energy prod. (Wh/km2/a) PV capacity per area (W/km2) Areas potentially suitable for PV systems (km2) Site assessment (solar atlas data, solar radiation (kWh/m²/a); open-land and settlements (roofs) Exclusion of non-suitable areas Nature conservation areas Exclusion of non-suitable built-up areas (i.e. non-suitable roofs) Transport, supply and communication infrastructure; very remote areas Areas technically not suitable (high slope and above certain altitude, etc.) Landscape, historic area, other non- usable land (glaciers, rivers, roads etc.) Areas potentially suitable for PV systems (km2) Priority areas for PV (km2), potentially installed capacity (W), potentially generated energy (Wh/a) and costs Energy policy analysis Economic assessment Performance Ratio done done CAPEX = Capital expenditure, OPEX = Operation expenditure, WACC = Weighted average cost of capital (debt, equity)
- 4. Contents 1. Levelized cost of electricity (LCOE) 2. Worked example: LCOE sensitivity of PV projects 3. Worked example: LCOE sensitivity of wind projects 4. Worked example: Effects of data uncertainty on the LCOE of PV 4
- 5. 1. LEVELIZED COST OF ELECTRICITY (LCOE) 5
- 6. Levelized Cost of Electricity (LCOE) • Calculates the average cost per unit electricity. LCOE takes into account the time value of money (i.e. capital costs). Where: • LCOE: Average Cost of Electricity generation in $/unit electricity • I0: Investment costs in $ • At: Annual total costs in $ in each year t • Qel: Amount of electricity generated • i: Discount interest rate in % • n: useful economic life • t: year during the useful life (1, 2, …n) 6
- 7. 2. LCOE SENSITIVITY OF PV PROJECTS Worked example: 7
- 8. Worked example – Grid-tied PV in Egypt • Project type: Grid-tied • Location at latitude: 20° North • Reference irradiation (GHI): 2,500 kWh/m²/a • Reference specific yield (P50): 1,880 MWh/MWp • System size: 10 MWp • Specific project CAPEX: 2.000.000 USD/MWp • Project annual OPEX: 1.5% of project CAPEX • Discount rate (WACC): 8% • Project duration: 30 years • Inverter replacements: 2 • Solar panel degradation: 0,7% p.a. (linear) 8
- 9. LCOE sensitivity (absolute) 9 Baseline LCOE: 123 USD/MWh
- 10. LCOE sensitivity (relative) 10 Baseline LCOE: 123 USD/MWh
- 11. 3. LCOE SENSITIVITY OF WIND PROJECTS Worked example: 11
- 12. Worked example – Grid-tied wind project Egypt (variation A) • Project type: Grid-tied wind • Location: Egypt / south-west of Cairo • Average wind speed @ 80m: 7.3 m/s • Wind distribution, shape parameter: 3.5 • Wind distr., scale parameter: 8.11 • Technical availability: 97% • Reference specific yield (P50): 3,202 MWh/MW (techn. Availability considered) • Capacity factor: 36.6% • System size: 8 MW (4 turbines) • Specific project CAPEX: 4.000.000 USD per turbine • Project annual OPEX: 3.0% of project CAPEX • Discount rate (WACC): 8% • Project duration: 20 years 12
- 13. LCOE sensitivity (absolute) – Wind speed only 13 Baseline LCOE: 87.6 USD/MWh
- 14. LCOE sensitivity (absolute) – other parameters 14 Baseline LCOE: 87.6 USD/MWh
- 15. Worked example – variation B: lower wind speed & lower shape parameter • Project type: Grid-tied wind • Location: Egypt / south-west of Cairo • Average wind speed @ 80m: 7.3 m/s 5.5 m/s • Wind distribution, shape parameter: 3.5 m/s 1.5 m/s • Wind distr., scale parameter: 6.11 • Technical availability: 97% • Reference specific yield (P50): 2,054 MWh/MW (techn. Availability considered) • Capacity factor: 23.5% • System size: 8 MWp (4 turbines) • Specific project CAPEX: 4.000.000 USD per turbine • Project annual OPEX: 3.0% of project CAPEX • Discount rate (WACC): 8% • Project duration: 20 years 15
- 16. LCOE sensitivity (absolute) – Wind speed only 16 Baseline LCOE: 136.6 USD/MWh
- 17. LCOE sensitivity (absolute) – other parameters 17 Baseline LCOE: 136.6 USD/MWh Shape parameter more sensitive!!!
- 18. Conclusions on sensitivities and for scenario development • Variations of the shape parameter of the Weibull distribution of wind can have very different effects depending on the chosen scenario In variation A (high wind, high shape factor), varying of the shape factor only had a very little effect on the LCOE. In variation B (lower wind, lower shape factor), varying of the shape factor had a considerable effect on the LCOE. Reason: the chosen wind turbine for the scenario has a power curve which operates better under stronger winds. The Weibull function produces a wind distribution where relative low wind speeds occur comparably often. It is crucial for wind scenario developments, to chose appropriate turbines for sites with different wind speeds and wind speed distributions. 18
- 19. Comparison of Weibull curves for variations A (left ) + B (right) 19
- 20. 4. EFFECTS OF DATA UNCERTAINTY ON THE LCOE OF PV Worked example: 20
- 21. Why data quality is so important • All data comes with uncertainties: Measurements are always subject to deviations, and , models used for predictions can never simulate what happens in reality. • It is obvious that the lower uncertainty is the more accurate predictions will be. This, in turn, will enable us to make better estimates. • In the following, we will demonstrate how good data (i.e. data with low uncertainties) will potentially help saving funds for PV Power Purchase Agreements. 21
- 22. Uncertainty assumptions • Low resolution NASA SSE data: +/- 13,7% • Average Meteonorm 7 data: +/- 7,5% • Best ground measurement at site: +/- 3,0% • Important note: Besides uncertainty of irradiation data, there is also uncertainty within the simulation model and nameplate capacity. However, the latter are comparably small so that we will, to keep the example simple, only look at resource uncertainty. In real-life, when it comes to detailed project development, one should always ask the project developer to provide information about his uncertainty assumptions. 22
- 23. Worked example – Grid-tied PV in Egypt • Project type: Grid-tied • Location at latitude: 20° North • Reference irradiation: 2500 kWh/m²/a • Reference specific yield (P50): 1880 MWh/MWp • System size: 10 MWp • Specific project CAPEX: 2.000.000 USD/MWp • Project annual OPEX: 1.5% of project CAPEX • Discount rate (WACC): 8% • Project duration: 30 years • Inverter replacements: 2 • Solar panel degradation: 0,7% p.a. (linear) 23
- 24. Exceedance probability (alternative view) 24 P50: 1880 MWh/MWp P90
- 25. LCOE depends on quality of meteo data 25
- 26. LCOE is key factor for PPA tariff calculation • Assuming a 10% premium on the LCOE as margin for IPP Best case: 128 USD/MWh +10% = 141 USD/MWh Worst case: 149 USD/MWh +10% = 164 USD/MWh Delta: 23 USD/MWh (incl. 10% premium) 26
- 27. Country sets a 5% PV goal by 2020 • Sample: Egypt • Total electricity demand 2012: 109 TWh (Source: indexmundi.com) • 5% of total: 5.45 TWh • PPA tariff difference: 23 USD/MWh • „Unnecessary“ payments in 2020: 5,450,000 MWh * 23 USD/MWh =125.4 Mio USD • PV power needed: 3,014 MWp (with best P90 value) 27
- 28. „Unnecessary“ payments due to inaccurate data • PV power needed by 2020: 3,014 MWp (with best P90 value) • Aviodable payments: 376 Mio USD 28
- 29. 29 © RENAC 2014 Irradiation on tilted plane (Wh/m²/a) Energy generation costs at specific site (€/Wh) Conversion horizontal solar radiation to optimally tilted plane Optimal tilt angle Energy output calculation Pre-conversion losses Conversion losses System losses (%) CAPEX OPEX WACC Life time Economic parameters (PV plant and grid connection) Annual energy prod. (Wh/km2/a) PV capacity per area (W/km2) CAPEX = Capital expenditure, OPEX = Operation expenditure, WACC = Weighted average cost of capital (debt, equity) Areas potentially suitable for PV systems (km2) Site assessment (solar atlas data, solar radiation (kWh/m²/a); open-land and settlements (roofs) Exclusion of non-suitable areas Nature conservation areas Exclusion of non-suitable built-up areas (i.e. non-suitable roofs) Transport, supply and communication infrastructure; very remote areas Areas technically not suitable (high slope and above certain altitude, etc.) Landscape, historic area, other non- usable land (glaciers, rivers, roads etc.) Areas potentially suitable for PV systems (km2) Priority areas for PV (km2), potentially installed capacity (W), potentially generated energy (Wh/a) and costs Energy policy analysis Economic assessment Performance Ratio done done done
- 30. Thank you very much for your attention! Volker Jaensch Renewables Academy (RENAC) Phone +49 30 52 689 58-85 [email protected] www.renac.de