#1ARCVERA RENEWABLES Golden - São Paulo - Cape Town - Bangalore New Tools Help Assure Success of South Africa Offshore Wind Energy Plans Long-Range Wake Losses, A Critical Consideration + Gregory S. Poulos, PhD, Principal Atmospheric Scientist, CEO Credit to Mark Stoelinga, PhD October 12, 2022 Photo: Henrik Krogh#2EXPERTISE THROUGHOUT THE PROJECT LIFECYCLE Prospecting Development Financing Construction Operations Repowering Atmospheric Science, Engineering, and Storage/Wind/Solar/Green H₂ Technological Consulting with Roots to the Very Origins of the U.S. Industry 9 Start 1978, named ArcVera March 2017 Origins of US renewables • California, post-oil embargo, 1978 start Deep historical project database Technology/engineering and expertise/experience used to cut risk Offices in US, Brazil, South Africa and India 1st January 1989 1st Wind farms in twenty five states of the U.S.A. 9GW Based in Golden, CO, US & Globally • Projects on 6 of 7 continents • 40+ person degreed team - 4 continents • • Senior staff 10+ years experience Work on thousands of storage/wind/solar projects globally OVER 70 Percent of all U.S. wind capacity evaluated 6 of 7 Continents with assessed and installed energy projects 60+ Types of wind turbine models reviewed 9 of 10 of the largest U.S. wind farms evaluated Wind farms in Puerto Rico, Nicaragua, Guatemala, and El Salvador Year of first financeable wind assessment in California Solar projects evaluated across North and South America Green H₂ projects underway, globally ARCVERA HYBRID • STORAGE • SOLAR WIND ⚫GH₂ RENEWABLES#3Background: Wind Turbine Wake Wind comes in fast & comes out the other side slower, up to about half as fast. Faster incoming wind Slower wind downwind. This is the wake. Less energy in the air for the next turbine downwind. Based on the CEWEX field experiment (and others) no substantive impact on farming or environment due to wake wind speed changes. Wakes and their recovery rate are a critical wind farm design factor. ARCVERA RENEWABLES#4Wind Turbines are Getting Much Bigger Onshore Wind Turbines Offshore Wind Turbines Capacity 17MW Capacity 2.5 MW Rotor Diameter 120m Hub Height 89m Rotor Diameter 174m Capacity 5.5 MW Hub Height 130m Capacity 6.0 MW Rotor Diameter 150 m Hub Height 103 m Rotor Diameter 250 m Hub Height 151 m 2019 Specific Power 221 W/m² From Renewable Energy World, 2021 2035 Specific Power 231 W/m² 2019 Specific Power 340 W/m² 2035 Specific Power 346 W/m² Taller turbines reach further into the atmosphere where it is more stable and wakes recover more slowly. Neutral and unstable atmospheres are more common near the surface, but overall are more rare than stable. Stability varies continuously. ARCVERA RENEWABLES#5Many in the industry think wakes recover to the upwind speed within 3-5 km or ~30-50 rotor diameters. What if the real answer is 10 times bigger? Then, "Add a zero". - It turns out that the state of the atmosphere - an inversion, for example – lengthens the distance wakes travel by 10x and this effect is not well captured in all commonly and currently used fast easy-to-use wake models, requiring a more sophisticated approach. Because long-range (> 30-50 RD) wake losses have not been considered significant they have not been validated in current wake loss models, leaving a gap in knowledge that increases energy production risk as we build more and more wind farms.#6How can we possibly understand something so complex? Weather Research and Forecasting model, WRF A state-of-the-art model from NCAR/NOAA Continuously models atmospheric physics and thermodynamics and naturally time-varying stability Has ocean interaction with the atmosphere Contains all the weather variables/turbulence Models wind turbine interaction with atmos. o Wind Farm Parameterization, WFP Thus, a solution is found with WRF-WFP. ARCVERA RENEWABLES#7ArcVera's Study of Long-Range Wakes Paper available with detailed references and validation Visit arcvera.com, navigate to "Resources", then "News and Publications". White paper: Estimating Long-Range External Wake Losses in Energy Yield and Operational Performance Assessments Using the WRF Wind Farm Parameterization Mark Stoelinga, et al. 2022 Key finding: WRF-WFP by far the most accurate long-range wake loss method. ARCVERA RENEWABLES#8Existing Wake Loss Tools Inadequate • Turbines are rapidly changing; tuned models can't keep up . They have low sensitivity to changing atmospheric stability • Tuned to close-in wind farm wakes from smaller turbines and with insufficient turbine size dependence, looking backward . BOTTOM LINE: Not validated or adequate for long-range external wake losses – a gap in knowledge that can easily - translate to billions of ZAR in lifetime lost revenue for a single offshore wind farm. USE WRF-WFP method to solve this problem. ARCVERA RENEWABLES#9Current Methods Found Unacceptably Inaccurate for Long Range Wakes How can WRF-WFP long-range wake loss solution be used for offshore wind development planning in South Africa? ARCVERA RENEWABLES#10Unique aspects of offshore environment around South Africa 15°00'E • 26°00'S- W 2000E 25°00°E 30°00'E 35°0″E 60-m depth contour is too close to shore along most* of South Africa coast • Wind flow is also mostly parallel to shore farm-to-farm waking common if not considered in farm placement! 30°00'S 35°00'S- Sadanha Bay 270° 90° 180° East London Ngqura Bort Elzabeth Town Mossel Bay From Rae, G., and G. Erfort (2020) 15°00'E 20°00'E 0° Richards Bay 270° 90° Durban 180° Legend Weighted Overlay - Suitability 30-38 % 38-45% 45-53% 53-60% 60-69 % 0° 68-75% 5% 75-83% 10% 270 90° 83-91% 91-98% 50% 180° 270° 90° Contour_-1000m 180° ° 100 200 Kilometers 400 RSA_EEZ Commercial Harbours 25.00€ 30°00′E 35°0'0"E -25°00's -30°cro's -35°00's ARCVERA RENEWABLES#11North Array A Hypothetical Cape Town Offshore Wind Farm Vredenburg Hopefield LanLangebaan Yzerfontein Swartwater Jakkalsfontein Darling BLOUBERGSTRAND MILNERTON Cape Town M Two 2.2 GW arrays, 25 km apart Each array has: • 16 rows of 9 turbines, 144 total 1 nautical mile (1.85 km) spacing in both directions IEA reference 15 MW turbine Rotor diameter = 240 m Hub height=150 m Array capacity: 2.2 gigawatts Two experiments: No turbines South Array turbines Assess affect of external wakes from South Array on North Array production 25 km Grotto Bay South Array ARCVERA RENEWABLES#12Animated Loop of Long-Range Wakes: 3 Days Long-Distance Wakes Offshore of Cape Town Using WRF Wind Farm Parameterization Current Frame: 2001-01-01 08:00 SAST Season: Summer Each frame is 10 minutes in simulated time. North Array Unwaked Speed in North Array: 0.00 m/s -5 -4 50 km South Array Langebaan -2 -1 -3 Waked Wind Speed Deficit (m/s) 0 Cape Town 0 1 ARCVERA RENEWABLES#13Snapshot of a Long-Range Wake During a full year of simulated wakes from the South Array, the average speed and energy deficits at the North Array are: • Speed: 1.7% Energy: 2.1% This energy loss would not be considered without WRF-WFP. Waked Wind Speed Deficit at 100 m Hub Height North Array Unwaked Speed in North Array: 7.84 m/s 50 km South Array 1400 SAST 01 July 2020 0.3 m/s (4%) Langebaan speed deficit at 75 km downwind M 1 m/s (13%) speed deficit at 25 km downwind 0 Cape Town T -5 -4 -2 -3 Waked Wind Speed Deficit (m/s) -1 0 1 ARCVERA RENEWABLES#14Summary Long-range wakes of large magnitude travel much further than expected, 300-500 rotor diameters (50-100 km or more). Long-range wakes are vastly underpredicted by traditional/current engineering wake loss models at these distances. • This new WRF-WFP based method excels. • This capability is available now and amounts to potentially many billions. of ZAR for a single project's lifetime revenue. • South Africa should carefully consider these impacts for accurate grid modeling. . Policy makers/developer/owners will now be able to act on best information and optimize offshore wind energy polygons. ARCVERA RENEWABLES#15Points to Finish . Bottom line: As on and offshore wind farm density increases, with ever larger rotor diameters, long-term green energy production risk increases greatly due to long-range wakes. We have demonstrated that the WRF-WFP works well. Use it to reduce project risk. Deeper Insight: Such analysis is critical for project optimization and hybrid project time series modeling of energy production and for grid stability/worst-case-scenario modeling and for state-of-charge/revenue modeling for green H, and battery projects. Recommendation: Considering the unique environment of wind and bathymetry offshore of South Africa, long-range wakes will be especially important here. Use WRF-WFP to study long time series of full-build effects on grid management and project performance. ARCVERA RENEWABLES