Torrild

Annual Energy Production Assessment

3 turbines Enercon

10. JANUAR 2020

1/26

Torrild Denmark

RECIPIENT Green Wind Denmark ApS Egå Havvej 21 DK-8250 Egå

Attn. Jørn Motzkus

DATE 10 January 2020

PREPARED BY EMD International A/S Niels Jernes Vej 10 DK- 9220 Aalborg T: + 45 96 35 44 44 E: [email protected]

PRINCIP A L CONSULTANT Madalina Calin EMD-DK

APPROVED BY Thomas Sørensen EMD-DK

DOCUMENT 200110_19179_A_MC_0

CLASSIFICATION Commercial in confidence

200110_19179_A_MC_0_ts2 EMD International A/S

2/26

D O C U M E N T R EVISIONS

REVISION DATE C H A N G E D CHAPTER DESCRIPTION OF CHANGE BY

0 10/01/2010 MC All

KEY TO DOCUMENT CLASSIFICATION Strictly Confidential: Recipients only Private and Confidential: For disclosure to individuals directly concerned within the recipient’s organisation. Commercial in Confidence: Not to be disclosed outside the recipient’s organisation Client’s Discretion: Distribution at the discretion of the client subject to contractual agreement. Published: Available to the general public.

LIABILITIES EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results caused by inaccuracy, limitations or malfunctioning of models or software used.

For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of the form of action, shall be limited to the total amount paid to EMD for the services provided as part of this consultancy job.

200110_19179_A_MC_0_ts2 EMD International A/S

3/26

Executive Summary

EMD has been assigned to carry out an annual energy production assessment on the Torrild wind farm project, Denmark. The project consists of three Enercon E82 2350 turbines with a hub height of 58.9m.

The assessment is conducted according to EMD’s standard procedure for AEP assessments in Denmark. This method implies the use of a downscaled mesoscale data and the calibration of the wind model with operational data from existing wind turbines. The downscaling methodology is moreover combined with a traditional WAsP vertical and horizontal extrapolation.

The annual energy production has been calculated and losses and uncertainties have been assessed. The resulting P50, P75 and P90 (production percentiles) are listed in table 1.

The production after losses (P50) is the sold electricity (at the billing meter).

Table 1. Annual Energy Production result.

G R O S S A E P W A K E L O S S AEP = G R O S S - WAKE MEAN WIND SPEED AT HUB [MWH/Y] I N T E R N A L - [ GWH/Y ] HEIGHT (FREE WIND SPEED), M/S EXTERNAL

15,511.3 6.00 14,586.8 6.4

Table 2. Annual Energy production, Capacity factor, Full load hours and uncertainty results at 20 years horizon N E T A E P 20Y HORIZON UNCERTAINTY

P - 50 C A P A C I T Y F L H P - 9 0 C A P A C I T Y F L H 1Y 2 0 Y [ M WH/YR] F A C T O R [ % ] [H] [ M WH/YR] F A C T O R [ % ] [H] [%] [%]

14,098 22.8 2,000 12,070 19.5 1,712 13.3% 11.2%

200110_19179_A_MC_0_ts2 EMD International A/S

4/26

Index

1 Project Description ...... 6 2 Description of the calculation method ...... 8 3 Reference turbines ...... 9 3.1 Available Data ...... 9 3.2 Normalized production procedure ...... 9 4 Wind Model ...... 12 4.1 Wind data ...... 12 4.2 Terrain data ...... 13 4.3 Wind model calibration ...... 14 5 AEP calculation ...... 18 5.1 Power curves ...... 18 5.2 Wake model ...... 18 5.3 Gross AEP ...... 18 6 Losses & bias ...... 19 6.1 Bias ...... 19 6.2 Wake losses ...... 19 6.3 Availability losses ...... 20 6.4 Electrical losses ...... 20 6.5 Environmental losses ...... 20 6.6 Curtailment losses ...... 20 7 Uncertainties ...... 22 7.1 Wind measurements ...... 22 7.2 Long term correction ...... 22 7.3 Year-to-year variation ...... 23 7.4 Future climate ...... 23 7.5 Vertical and horizontal extrapolation ...... 23 7.6 Power curve uncertainty ...... 23 7.7 Uncertainty on losses ...... 23 8 References ...... 24 Appendix A – Gross AEP calculation ...... 25 Appendix B – Net AEP calculation ...... 26

List of Figures

200110_19179_A_MC_0_ts2 EMD International A/S

5/26

Figure 1 Location (orange cross) of the Torrild wind farm. (Open Street maps) ...... 7 Figure 2 Location of the turbines of the Torrild wind farm project (red symbol) (map: Geodatastyrelsen) ...... 7 Figure 3: Monthly windiness index from 1999 to 2019...... 9 Figure 4: EMD-ConWx and EMD Europe (ERA5+) long-term yearly windiness index...... 10 Figure 5: Linear regression between ideal production and EMD Europe (ERA5+) energy index. Example for a 900kW Neg Micon-Torrild turbine...... 10 Figure 6 Correlations between the production data of the reference turbines and the 2 indices available...... 11 Figure 7 Location of the mesoscale node (orange symbol) and the turbines (red symbol) (map: Geodatastyrelsen) ...... 12 Figure 8 Energy rose at the EMD Europe (ERA5+) data point at 58.9m ...... 13 Figure 9 Input of forest areas used for the calculation of the displacement height for the 20year horizon situation...... 14 Figure 10. Location of reference turbines within 13 km radius of Torrild wind farm...... 15

List of Tables

Table 1. Annual Energy Production result...... 3 Table 2. Annual Energy production, Capacity factor, Full load hours and uncertainty results at 20 years horizon .. 3 Table 3 List of Torrild turbines with operational power modes. All coordinates are UTM ETRS89, Zone 32...... 6 Table 4 List of reference turbines ...... 15 Table 5 Reproduction of actual normalized production by the wind model ...... 16 Table 6 Gross production result ...... 18 Table 7 Summary of bias and losses ...... 19 Table 8 Summary of uncertainties ...... 22

200110_19179_A_MC_0_ts2 EMD International A/S

6/26

1 Project Description

The Torrild wind farm project is located in the municipality of , in Midtjylland Region, in Denmark. The project consists of 3 Enercon E82 2.35MW wind turbines. The turbine location and details can be found in the following table. Two turbines run in level 0 OM0-2350kW while the third runs level 2-2000kW. The entire wind farm is limited to 6.1MW production. Table 3 List of Torrild turbines with operational power modes. All coordinates are UTM ETRS89, Zone 32.

H U B N O M I N A L ELEVA - O P E R A T I O N NR. TYPE H E I G H T P O W E R EASTING NORTHING T I O N A G L MODE [M] [MW] [M]

ENERCON E-82 1 58.9 2.0 565817 6202673 80 Level 2 E4 2350

ENERCON E-82 2 58.9 2.35 566170 6202673 79.8 Level 0 E4 2350

ENERCON E-82 3 58.9 2.35 566524 6202673 75.5 Level 0 E4 2350

200110_19179_A_MC_0_ts2 EMD International A/S

7/26

Figure 1 Location (orange cross) of the Torrild wind farm. (Open Street maps)

Figure 2 Location of the turbines of the Torrild wind farm project (red symbol) (map: Geodatastyrelsen)

200110_19179_A_MC_0_ts2 EMD International A/S

8/26

2 Description of the calculation method

Traditionally site-specific measurements are not conducted in Denmark. Instead a nationwide wind atlas has been developed, calibrated with actual, long term adjusted production from existing wind farms across Denmark. Due to the uncomplicated nature of Danish terrain and the very extensive network of existing turbines combined with the ready availability of production data, this approach has by far proved superior or at least comparable to on-site measurements. In recent years, this approach has evolved through the use of the downscaling technique [1]. The methodology uses a long-term mesoscale dataset calibrated with actual, long term adjusted production from nearby existing wind farms. The first step of this approach uses a downscaling technique [1]. The EMD-ConWx mesoscale dataset, already calculated using a fairly crude terrain description (mesoscale terrain), is processed by WAsP [2]. The downscaling becomes particularly important in coastal areas where the mesoscale terrain map may be too crude to catch the sea-land transition. The WasP process cleans the mesoscale terrain effects from the virtual mast and introduces the microscale terrain effects from the location of the new and neighboring turbines. At the same time, a complete wind dataset is created at the hub height of the new turbines. In this case at the height of 58.9m. The downscaling method has the advantage of being applied in the time domain [3]. That means that every single data record in the mesoscale dataset is transferred by WAsP, eventually creating time series of transferred data at the selected hub height. The second step consist of a calibration. Due to the nature of mesoscale data which is often offset, the downscaled mesoscale wind speeds, at the new created height, is calibrated with normalized production data from existing turbines. The result is a scale factor which is applied to the entire wind speed datasets. Calibration is done against the long term adjusted production. This methodology is explained in chapter 3.1. The wind direction of the mesoscale data has proven to be quite precise, thus no calibration is applied for wind direction. This process is an iterative one and represents a combination of downscaling and traditional WAsP vertical and horizontal extrapolation. Downscaling is only used to generate a time series at the EMD-ConWx data node at a height representative of the new turbines. Thereafter the traditional WAsP practice takes over with a wind statistic created from the downscaled height. This is because the calibration points (the operating wind turbines) are usually at a different hub height from that of the new wind turbines and vertical modelling can be done with better confidence using the classic wind statistics (the wind atlas) model than with a time varying calculation. Once the wind statistics is able to reproduce the normalized production data from reference turbines, the model is considered calibrated.

200110_19179_A_MC_0_ts2 EMD International A/S

9/26

3 Reference turbines

3.1 Available Data

EMD has considered data from 12 exiting turbines, in a radius of 13km from the site. The production data was gathered from The Danish Energy Agency’s database on Danish turbines [4] (da. “Stamdataregisteret”) as sold production figures. EMD has processed the data and calculated the long term adjusted, annual average production for each turbine. The long-term basis is a 20 year time series of mesoscale data as described in section 4.1. The periods where the production data shows a decreasing trend are disregarded in order to avoid any bias in the calibration of the model. Decreasing trend is indeed quite commonly observed for old turbines due to the degradation of their performance. In the case were a given wind farm is becoming exposed to the wake from a new wind farm, it is decided to keep either the period before or after the new turbines were build and model this scenario consequently. The database holds no information about availability or other external losses. In the filtering procedure of production data outliers are removed, but we assume that on average 1% of availability loss and 1% of electrical loss (from the turbine transformer) remains embedded in the data. A calibration with this data basis will therefore have to be adjusted with a corresponding bias of +2%.

3.2 Normalized production procedure

EMD has processed the monthly production of the reference turbines. The normalized production of the reference turbines is found through a linear regression analysis of the production against a windiness index. A monthly wind index is generated from two data sets, extracted from a grid point at a representative location of the project: • EMD ConWx mesoscale data (20 years, September 1999 – August 2019), hourly data • EMD Europe (ERA5+) (20 years, September 1999-August2019), hourly data. The monthly wind index corresponds to the monthly production relative to the 20-year average. All datasets have been checked for trends with Mann-Kendall test. The test returned 0.4 for EMD ConWx, and 0.3 for EMD Europe (ERA5+) datasets, meaning that none of these datasets experience a significant trend.

Figure 3: Monthly windiness index from 1999 to 2019.

200110_19179_A_MC_0_ts2 EMD International A/S

10/26

This can be seen more clearly in the yearly version of the above graph, as shown below.

Figure 4: EMD-ConWx and EMD Europe (ERA5+) long-term yearly windiness index. A regression analysis is then conducted for each wind turbine (example in Figure 5) and each index.

Figure 5: Linear regression between ideal production and EMD Europe (ERA5+) energy index. Example for a 900kW Neg Micon-Torrild turbine. On the figure above, the white points are disregarded because of their too large deviation resulting from, presumably, a very low availability. The correlation of the monthly production data with the monthly wind energy index can be seen in the figure below. The correlation (R2) is in average for all turbines 0.94 with EMD Europe (ERA5+) index and 0.92 with EMD ConWx. Wind turbines with a correlation lower than 0.80 are excluded from the analysis.

200110_19179_A_MC_0_ts2 EMD International A/S

11/26

Figure 6 Correlations between the production data of the reference turbines and the 2 indices available.

The final wind energy index selected is the one obtained from EMD Europe (ERA5+) data since it has the highest correlation, the highest Mann-Kendall test figure and a 20 years data availability. This index is selected for the long-term correction of the ideal production of the reference turbines. The normalized production is further used to calibrate the wind model, as described in section 2.1.

200110_19179_A_MC_0_ts2 EMD International A/S

12/26

4 Wind Model

As described in the previous section, the wind model is a downscaling of mesoscale data using mesoscale terrain and microscale terrain. The model is subsequently calibrated with reference turbine production and a traditional WAsP methodology is used to extrapolate the wind model to hub height of each turbine.

4.1 Wind data

The mesoscale data used in the calculation are the EMD Europe (ERA5+) data [5], developed and produced by EMD. The data node is located close to the wind farm (Figure 7) yet sufficiently removed from the coastline to avoid coastal problems with meso scale terrain. EMD Europe (ERA5+) wind energy index is used for the long-term normalization of the production data of the reference turbines, and it will be further used in the wind model calibration methodology. Data are extracted at six heights from 10 to 200 m for the period September 1999 to August 2019 (20 years). The data has been checked for long term consistency. The mesoscale data are then downscaled to 58.9 m height which is the hub height of the planned wind turbines. Based on the calibration process described in section 4.3, the wind speeds are scaled with a factor of 0.8922. As described in section 4.3, this final scaling factor is obtained in the AEP calculation allowing for the best match between actual and calculated production of the selected reference turbines. The scaling factor corresponds to a decrease of 10.78% of the mesoscale wind speed data. Used at first for the reference turbines (calibration) following a traditional WAsP calculation, the scaled data set is secondly used similarly for the new turbines, the subject of this AEP study. The 20-year mean wind speed at 58.9m at the position of the EMD Europe (ERA5+) data point is thus 6.28m/s. The main wind directions (energy wise) are WSW and W.

Figure 7 Location of the mesoscale node (orange symbol) and the turbines (red symbol) (map: Geodatastyrelsen)

200110_19179_A_MC_0_ts2 EMD International A/S

13/26

Figure 8 Energy rose at the EMD Europe (ERA5+) data point at 58.9m

4.2 Terrain data

4.2.1 Roughness and elevation

The wind farm is located in a flat landscape bordered to the North East and South by large forest plantations. The terrain is relatively opened with a combination of open fields and patches of forest and wind breaks. The distance to the East coast and to the large open water area is about 13km East and 10km South to . The terrain data from the mesoscale model is used for first half of the transfer process. Detail and quality level are discussed in [5] . The microscale terrain is composed of a roughness map and an elevation map. The roughness map extracted from the Corine land cover 2018 dataset [F] is rasterized roughness map with a resolution of 100 m. The roughness values of the map are calibrated with information from orthophoto maps. A background roughness class 1.8 is used. The elevation map is the 5m contour map provided by Geodatastyrelsen [6]. Both maps extend beyond the radius around the site and the reference turbines normally recommended.

4.2.2 Forest

The largest patches of forest on the site will cause a displacement height at the reference turbines and the Torrild wind turbines. The average height of the trees at the current situation and in 25 years horizon have been estimated based on surface data from Ustun windProsper maps, orthophotos, Google Street view and the data

200110_19179_A_MC_0_ts2 EMD International A/S

14/26

base on Trees in Denmark (www.Træ.dk). From the height input, the internal windPRO displacement height calculator can then estimate the sector wise displacement height for each turbine (Appendix A).

Figure 9 Input of forest areas used for the calculation of the displacement height for the 20year horizon situation.

4.3 Wind model calibration

Within the area with a radius of 13 km, a number of separate wind farm clusters can be found and used as reference turbines. This excludes very small and household turbines. Predictions have been run on turbines within this capacity interval: 600kW-2.2MW. The smaller turbines have the advantage of many months of available data, while the taller and newer turbines are interesting for the calibration since they are in the same range (of rated power and rotor diameter) as the turbines from Torrild. However, such turbines are also prone to operate in reduced modes, which is unknown to us in details. Another drawback is that these tall turbines can be subjected to grid regulation (starting in 2015), which is also an information not publicly available. That is why, the success of the prediction of the annual energy production of these turbines is arbitrary. The result of predicting the reference turbines are presented in Figure 10 and Table 4.

200110_19179_A_MC_0_ts2 EMD International A/S

15/26

Figure 10. Location of reference turbines within 13 km radius of Torrild wind farm.

Table 4 List of reference turbines

CLUSTER UNITS KW/ R O T O R H U B Z D I S T A N C E ID WTG [M] H E I G H T [M] [KM] [M]

3 x 600 kW Bonus - Park_1 3 600 44 45 81.0 3.5

3 x 900 kW NEG Micon - Park_2 Torrild 3 900 52 49 79.3 1.8

2 x 2200 kW Vestas Wind Park_3 Systems - Trust 2 2200 100 75 59.0 6.2

200110_19179_A_MC_0_ts2 EMD International A/S

16/26

3 x 750 kW NEG Micon - Park_4 Blegind 3 750 48.2 45 60.7 11.7

1 x 660 kW Vestas Wind Park_5 Systems - Gersdo 1 660 47 50 13.3 12.4

The actual long-term production of the existing turbines is firstly calculated: • using the EMD Europe (ERA5+) wind index • excluding unreliable data points or periods. The production of the reference turbines is then predicted using the wind model and the results are compared to the annual long-term production. If the wind model successfully predicts the turbines, the model is calibrated. The ratio in percent of the actual long-term production to the predicted production is calculated and presented below.

Table 5 Reproduction of actual normalized production by the wind model

F I N A L ID CLUSTER USED FOR THE CALIBRATION PROD/PRED

Yes, slightly underestimated maybe due Park_1 3 x 600 kW Bonus - Hundslund 102% to the assumptions for displacement height.

Yes, slightly underestimated maybe due 3 x 900 kW NEG Micon - Park_2 103% to the assumptions for displacement Torrild height.

Yes, tall wind turbines. Overestimation 2 x 2200 kW Vestas Wind Park_3 91% possibly due to unknown noise Systems - Trust curtailments or grid regulation.

Yes, slightly underestimated by the 3 x 750 kW NEG Micon - Park_4 104% model, possibly because the climate is Blegind more affected by the Solbjerg Sø

1 x 660 kW Vestas Wind No, wind farm in a different wind climate, Park_5 91% Systems - Gersdo next to the east coast

Comments on the results: The aim of these predictions is to be able to predict the average of all reference turbines with a ratio of AEP Produced/Predicted of 100% (third column in Table 5). The average ratio of AEP Produced/Predicted obtained from the four selected Parks is 100%. The selected wind farms are the ones stated with “Yes” in the Column “Used for the calibration” in Table 5.

200110_19179_A_MC_0_ts2 EMD International A/S

17/26

Among the 5 reference wind farms, 4 have been selected. The most important reference wind farms are the ones with newer and “tall” turbines and the ones located in similar environment. These wind farms are underlined in Table 5. For example, Park 2 consisting of three 900 kW NEG Micon turbines with a hub height of 49m, 1.6 km from Torrild and in a similar environment) are underpredicted by 3% which is acceptable considering the forest is assumed to have had a fixed height during the entire production period of these turbines. At this level the wind model is considered calibrated and representative for the climate observed on site.

200110_19179_A_MC_0_ts2 EMD International A/S

18/26

5 AEP calculation

The calculation of Annual Energy Production is done in WindPRO using the calibrated wind model described above, the layout submitted by the client and the N.O. Jensen wake model.

5.1 Power curves

The wind farm layout is described in section 1. The wind farm operates in level 0 (2.35MW) for two of the turbines and level 2 (2.0MW) for the third turbine. The air density is obtained from a weather station from Tirstrup and transferred to the hub height of the turbines resulting in a turbine specific air density of 1.239 kg/m3. The power curve is modified by windPRO to reflect this air density.

5.2 Wake model

Wake loss is calculated in same process as the AEP calculation. The wake model used is the N.O. Jensen model. The wind farm is located in rather open farmland and thus the wake decay constant of 0.075 is used as representative for the turbulence level at 58.9m over land.

5.3 Gross AEP

The AEP calculation uses the 20 years time series at 58.9m height described in section 4.1. This time series is used to calculate a WAsP wind statistic calibrated to match the existing turbine production. The displacement height is calculated according to the current situation. The wind statistics is then used to predict the wind climate at each turbine height and location. The power curve is eventually applied to provide the gross annual output for each turbine. In addition, wake losses are calculated using the wake model described in section 5.2. The external wake losses due to neighboring wind farms are estimated. The contribution of the other existing wind farms (13km away) is very low, corresponding to 0.3%. No expected wind farm project close enough to Torrild wind farm is known. The results of the Gross AEP are presented in Table 6 and in the calculation report in Appendix A. The results include external losses. The gross result is calculated with the displacement height assumed for the current situation (cf. section 4.2.2). Table 6 Gross production result

G R O S S A E P W A K E L O S S AEP = G R O S S - WAKE MEAN WIND SPEED [MWH/Y] I N T E R N A L - [ GWH/Y ] AT HUB HEIGHT EXTERNAL (FREE WIND SPEED), M/S

15,511.3 6.00 14,586.8 6.4

200110_19179_A_MC_0_ts2 EMD International A/S

19/26

6 Losses & bias

Bias and losses have been assessed or assumed according to best practices. They are described in detail below. Losses are listed for each turbine in Appendix B and summarized below. The production after losses (P50) is the expected production (at the billing meter).

Table 7 Summary of bias and losses

BIAS %

Included electrical and availability losses in production data +2.00

LOSSES 2 0 Y [ % ]

Wake loss (internal and external) 6.00

Availability loss 3.00

Turbine downtime 0.60

Electrical loss 1.00

Curtailment losses 0.73

Total loss 10.9

6.1 Bias

The wind model has been calibrated against the available production data from the reference turbines. However, this data contains a part of electrical losses and availability losses. It is generally assumed that these losses are equivalent to 1% for electrical losses and 1 % for availability losses. Consequently, the calculated production should be 2% higher in energy. A bias of +2% on production is applied.

6.2 Wake losses

The presence of one wind turbine in front of another will have a detrimental effect on the performance of the downwind wind turbine. This is due to the velocity deficit and the additional turbulence in the wake of the upwind turbine. The extent of the wake depends among other things on the thrust coefficient of the wind turbines and the ambient turbulence intensity. Wake effects are estimated for each wind turbine for each directional sector using the N. O. Jensen model with a wake decay constant of 0.075. The wake loss within the wind farm resulted in 5.7%. The external wake loss due to existing wind farms is 0.3%, resulting so in a total of 6% losses.

200110_19179_A_MC_0_ts2 EMD International A/S

20/26

6.3 Availability losses

Availability is given as a percentage of time/energy. Time and production are assumed to be equivalent. A warranted availability on a turbine level is assumed to 97% over the lifetime (20 years) [7]. The loss due to the availability of the Balance of Plant (BoP) of Danish turbines is 0% under the assumption that the BoP of the new turbines will have the same availability in average than the availability of the BOP of all reference turbines. The availability of the BoP of the reference turbines is already included in the production data since the production of reference turbine is reported at the metering point. The grid availability in Denmark is 100%, so no losses are to be expected on that account. Turbine downtime as a result of force majeure events, scheduled maintenance, and repair delays due to high winds or lack of spare parts, which are typically not covered under traditional warranties is estimated to 0.6%.

6.4 Electrical losses

As production is sold at the turbine the only electrical loss to take into account is the loss from the internal transformer. This is assumed to be 1% of production. The own consumption of the turbine can be calculated from SCADA data (minimum 6 month). It has not been possible to calculate it here.

6.5 Environmental losses

DEGRADATION The performance of wind turbines can be affected by blade degradation which includes the accretion of dirt, salt or insects, which may be washed off by rain from time to time, as well as physical degradation of the blade surface over prolonged operation. Mechanical deterioration of the blade [8] and drive train will compromise the general performance of the wind turbine. The degradation will change in time [9]. However, since the wind model is calibrated on already degraded turbines, it is assumed this loss is already factored in the model, thus no further degradation losses are applied.

ICING According to EMD’s experience on post construction analysis of existing wind farms located in similar environment in Denmark, there is no indication that losses due to icing events are to be expected on the site of Torrild.

6.6 Curtailment losses

GRID CURTAILMENT The curtailment is usually applied by the grid operator whenever the grid needs to be balanced or when the optimization of the electricity market makes it more favorable to shut down the turbines. EMD will not estimate a loss due to grid curtailment, as the lack of details make it difficult to calculate. There is no specification on how the grid curtailment is applied or when. In any case the grid curtailment is financially compensated every year.

200110_19179_A_MC_0_ts2 EMD International A/S

21/26

POWER LIMITATION

EMD estimated a loss due to power curtailment to 6.1MW. The loss calculation is based on hourly production values. Every hour time step with a calculated power greater than 6.1MW is curtailed to 6.1MW. The loss represents the difference between the total estimated production of the wind farm and the total curtailed production to a power of 6.1MW, resulting so in a loss of 0.73%.

NOISE CURTAILMENT No noise curtailment is imposed on the turbines of the wind farm. The owner of the wind farm has informed that no noise curtailment is applied. SHADOW FLICKER CURTAILM ENT No shadow curtailment is imposed on the turbines of the wind farm. The owner of the wind farm has informed that no shadow curtailment is applied.

200110_19179_A_MC_0_ts2 EMD International A/S

22/26

7 Uncertainties

Uncertainty is assessed as either uncertainty on wind speed or on production output. In case of wind speed uncertainty, the uncertainty is translated to production uncertainty using a sensitivity factor. The sensitivity factor depends of the turbine type and the mean wind speed. For this calculation, the sensitivity factor equals 2.4. The combined uncertainties are listed in the Loss and Uncertainty calculation report in Appendix B. Summaries are presented below.

Table 8 Summary of uncertainties

7.1 Wind measurements

The uncertainty on the wind speed is derived from the scatter on the reproduction of reference turbine production. The selected method has resulted in a rather large scatter on the reproduction degree. The uncertainty is 3.0% on the wind speed.

7.2 Long term correction

As the data used are already 20 year time series no long term correction of data is necessary. However, the model is calibrated with reference turbines and their production are long term corrected with the MERRA2 wind index. We estimate an uncertainty on this correction to 3.0% on the production.

200110_19179_A_MC_0_ts2 EMD International A/S

23/26

7.3 Year-to-year variation

The year to year variability of the wind speed in Denmark is estimated to 3.1%.

7.4 Future climate

The future climate is the potential difference in wind speed of the next 20 years from the past period considered in the wind study. Northern Europe is subject to long wave oscillations meaning that a 20 year operation period can be quite different from the very long term average. We estimate that for the wind speed regime in question this uncertainty is 1.5% on wind speed.

7.5 Vertical and horizontal extrapolation

The wind model is calibrated with reference turbines surrounding the site in a radius of 13km. Some wind turbines have hub heights similar to the hub height from Torrild. The uncertainty on the vertical extrapolation is set to 1% and horizontal extrapolation is 1.3% each on the wind speed. In addition to these, the uncertainty on the assumption of the displacement height is estimated to 2% on the production.

7.6 Power curve uncertainty

The uncertainties related to power conversion are - in contrast to the above - expressed an uncertainty of energy production.

POWER CURVE The site conditions vary from the test conditions during the certification process of the WTG. It is widely acknowledged that the performance depends on site-specific conditions like turbulence, shear and inflow. Therefore, an uncertainty is assigned to site-specific factors as well as sub-optimal operational settings like yaw error. The uncertainty is 2.9% [13]. METERING The accuracy of the fiscal substation energy meter is typical utility meter accuracy of 0.3%.

7.7 Uncertainty on losses

The uncertainty on each loss component is assessed individually as an uncertainty on the lost production. The site and layout are within the normal envelope of the wake model used. The standard uncertainty of the wake model is 20%. The uncertainty on turbine availability is assumed to be 50%. The electrical loss is a standard value for the loss in the internal transformer. The uncertainty assumed is 25%. The uncertainty on all remaining losses is set to 25%.

200110_19179_A_MC_0_ts2 EMD International A/S

24/26

8 References

[1] B. et.al., "Wind-Climate Estimation Based on Mesoscale and Microscale Modeling: Statistical–Dynamical Downscaling for Wind Energy Applications," DTU Wind Energy, 2014. [2] T. I. a. P. E.L, "European Wind Atlas," Risø National Laboratory. Denmark, 1989. [3] E. International, "Time Varying AEP Park Calculation. windPRO Energy (manual), section 3.4.1.2.," 2015. [4] E. o. D. Styrelsen, "Stamdataregisteret for vindmøller - https://ens.dk/". [5] E. International, "https://help.emd.dk/mediawiki/index.php?title=EMD-WRF_Europe%2B," 2019. [6] Geodatastyrelsen, "http://download.kortforsyningen.dk/," 2018. [7] Vestas, "Aftale om Service og Driftstilgængelighed (AOM5000)," 2017. [8] N. Gaudern, "A practical study of the aerodynamic impact of a wind turbine blade leading edge erosion," Journal of Physics: Conference Series 524 (2014), 2104. [9] M. Wilkinson, "Long-term Performance of Wind Farms," EWEA Technical Workshop Malmö, 2014. [10] V. W. S. A/S, "General Specification V126-3.6 MW High Torque ”, Document no.: 0056-4782 V02,," 2016. [11] B. S. Søndergård, "Grønkær vindmøllepark, Miljømålinger-Ekstern støj," 2018. [12] Vestas, "4130-Shadow Control Files Groenkaervej.zip," 2018. [13] M. Young, "Power Performance Testing: Truly Useful of Just Box Ticking," in EWEA Power Performance Workshop, Lyon, France, 2012.

200110_19179_A_MC_0_ts2 EMD International A/S

25/26

Appendix A – Gross AEP calculation

200110_19179_A_MC_0_ts2 EMD International A/S Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Main Result Calculation: 3x E82 E4 2.35MW 58.9Hh Wake Model N.O. Jensen (RISØ/EMD)

Calculation performed in UTM (north)-WGS84 Zone: 32 At the site centre the difference between grid north and true north is: 0.9°

Power curve correction method New windPRO method (adjusted IEC method, improved to match turbine control) Air density calculation method Height dependent, temperature from climate station Station: TIRSTRUP Base temperature: 7.8 °C at 25.0 m Base pressure: 1013.3 hPa at 0.0 m Air density for Site center in key hub height: 78.0 m + 59.0 m = 1.239 kg/m³ -> 101.1 % of Std Relative humidity: 0.0 %

Wake Model Parameters Terrain type Wake decay constant DTU default onshore 0.075

Displacement heights fromUstund

Wake calculation settings Angle [°] Wind speed [m/s] start end step start end step 0.5 360.0 1.0 0.5 30.5 1.0

Wind statistics DK EmdEuropeEra5_N55.974922_E010.049438 (2) - J Scale 58.90 m.wws(C) OpenStreetMap contributors, Data OpenStreetMap and contributors, ODbL WAsP version WAsP 11 Version 11.06.0028 Scale 1:20,000 New WTG Site Data Key results for height 59.0 m above ground level Terrain UTM (north)-WGS84 Zone: 32 Easting Northing Name of wind distribution Type Wind energy Mean wind speed Equivalent roughness [kWh/m²] [m/s] A 565,493 6,203,779 Site data: WAsP EMD Europe WAsP (WAsP 11 Version 11.06.0028) 2,196 6.3 1.9

Calculated Annual Energy for Wind Farm Specific results¤) WTG combination Result GROSS (no loss) Wake loss Capacity Mean WTG Full load Mean wind speed PARK Free WTGs factor result hours @hub height [MWh/y] [MWh/y] [%] [%] [MWh/y] [Hours/year] [m/s] Wind farm 14,586.8 15,511.3 6.0 23.6 4,862.3 2,069 6.4 ¤) Based on wake reduced results and any curtailments. Calculated Annual Energy for each of 3 new WTGs with total 7.1 MW rated power WTG type Power curve Annual Energy Links Valid Manufact. Type-generator Power, Rotor Hub Displacement Creator Name Result Wake Free rated diameter height height loss mean wind speed [kW] [m] [m] [m] [MWh/y] [%] [m/s] 1 A Yes ENERCON E-82 E4-2,350 2,350 82.0 58.9 Sector wise EMD Level 2 - official - 2000kW s - 04/2016 4,975.4 3.0 6.40 2 A Yes ENERCON E-82 E4-2,350 2,350 82.0 58.9 Sector wise EMD Level 0 - 2350kW - official - OM 0 - 2350kW - 06/2015 4,868.3 7.9 6.40 3 A Yes ENERCON E-82 E4-2,350 2,350 82.0 58.9 Sector wise EMD Level 0 - 2350kW - official - OM 0 - 2350kW - 06/2015 4,743.1 7.0 6.31

Annual Energy result includes shown losses. Additional losses and uncertainty must be considered for an investment decision.

WTG siting UTM (north)-WGS84 Zone: 32 Easting Northing Z Row data/Description [m] 1 New 565,817 6,202,673 80.0 ENERCON E-82 E4 2350 82.0 !O! hub: 58.9 m (TOT: 99.9 m) (2) 2 New 566,170 6,202,673 79.8 ENERCON E-82 E4 2350 82.0 !O! hub: 58.9 m (TOT: 99.9 m) (3) 3 New 566,524 6,202,673 75.5 ENERCON E-82 E4 2350 82.0 !O! hub: 58.9 m (TOT: 99.9 m) (4)

*) Included in wake losses is influence from 12 WTG(s) in the neighborhood, which has status as "Reference WTGs", see separate report to identify these. windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 1 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Reference WTGs Calculation: 3x E82 E4 2.35MW 58.9Hh Wake Model N.O. Jensen (RISØ/EMD)

Calculation performed in UTM (north)-WGS84 Zone: 32 At the site centre the difference between grid north and true north is: 0.9°

Power curve correction method New windPRO method (adjusted IEC method, improved to match turbine control) Air density calculation method Height dependent, temperature from climate station Station: TIRSTRUP Base temperature: 7.8 °C at 25.0 m Base pressure: 1013.3 hPa at 0.0 m Air density for Site center in key hub height: 78.0 m + 59.0 m = 1.239 kg/m³ -> 101.1 % of Std Relative humidity: 0.0 %

Wake Model Parameters Terrain type Wake decay constant DTU default onshore 0.075

Displacement heights fromUstund

Wake calculation settings Angle [°] Wind speed [m/s] start end step start end step 0.5 360.0 1.0 0.5 30.5 1.0

Wind statistics DK EmdEuropeEra5_N55.974922_E010.049438 (2) - J Scale 58.90 m.wws(C) OpenStreetMap contributors, Data OpenStreetMap and contributors, ODbL

WAsP version WAsP 11 Version 11.06.0028 Scale 1:400,000 New WTG Existing WTG Site Data Key results for height 59.0 m above ground level Terrain UTM (north)-WGS84 Zone: 32 Easting Northing Name of wind distribution Type Wind energy Mean wind speed Equivalent roughness [kWh/m²] [m/s] A 565,493 6,203,779 Site data: WAsP EMD Europe WAsP (WAsP 11 Version 11.06.0028) 2,196 6.3 1.9

Calculated Annual Energy for reference WTGs Specific results Specific results Calculated GROSS (no loss) Wake loss Capacity Mean WTG Full load Mean wind speed Actual wind corrected energy Goodness Factor prod. without Free WTGs factor result hours @hub height new WTGs [MWh/y] [MWh/y] [%] [%] [MWh/y] [Hours/year] [m/s] [MWh/y] [%] 28,784.8 29,882.4 3.8 27.8 2,398.7 2,437 6.2 27,642.0 96.0 Calculated Annual Energy for each of 12 reference WTGs with total 11.8 MW rated power WTG type Power curve Links Valid Manufact. Type-generator Power, Rotor Hub Displacement Creator Name Calculated Actual Goodness rated diameter height height prod. wind Factor without corrected new energy WTGs [kW] [m] [m] [m] [MWh/y] [MWh/y] [%] 4 A No VESTAS V47-660/200 660 47.0 50.0 Sector wise EMD Level 0 - calculated- - 07-2001 1,405.8 1,274.1 91 5 A No BONUS MK IV-600/120 600 44.0 45.0 Sector wise EMD Man. 03/02 1,139.2 1,121.7 98 6 A No BONUS MK IV-600/120 600 44.0 45.0 Sector wise EMD Man. 03/02 1,073.8 1,107.9 103 7 A No BONUS MK IV-600/120 600 44.0 45.0 Sector wise EMD Man. 03/02 1,046.5 1,085.7 104 8 A No NEG MICON NM52/900-900/200 900 52.0 49.0 Sector wise EMD Man. 01-2002 with 2 stall strips 1,767.4 1,842.1 104 9 A No NEG MICON NM52/900-900/200 900 52.0 49.0 Sector wise EMD Man. 01-2002 with 2 stall strips 1,740.1 1,796.4 103 10 A No NEG MICON NM52/900-900/200 900 52.0 49.0 Sector wise EMD Man. 01-2002 with 2 stall strips 1,663.0 1,700.4 102 11 A No VESTAS V100-2,200 2,200 100.0 75.0 Sector wise EMD Level 0 - Calculated - Mode 0 - 11-2016 7,474.8 6,953.0 93 12 A No VESTAS V100-2,200 2,200 100.0 75.0 Sector wise EMD Level 0 - Calculated - Mode 0 - 11-2016 7,591.2 6,704.2 88 13 A No NEG MICON NM48/750-750/200 750 48.2 43.6 Sector wise EMD Windtest/Man. 09-1999 1,305.6 1,352.2 104 14 A No NEG MICON NM48/750-750/200 750 48.2 45.0 Sector wise EMD Windtest/Man. 09-1999 1,269.3 1,352.2 107 15 A No NEG MICON NM48/750-750/200 750 48.2 45.0 Sector wise EMD Windtest/Man. 09-1999 1,308.2 1,352.2 103

WTG siting UTM (north)-WGS84 Zone: 32 Easting Northing Z Row data/Description Production source Statistical basis for normalized production: [m] [Months] 4 576,896 6,197,171 13.3 570715000000023279: 660 kW Vestas Wind Systems A/S - Gersdorffslund EmdEuropeEra5: 10.05E-55.97N-75m 196

To be continued on next page... windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 2 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Reference WTGs Calculation: 3x E82 E4 2.35MW 58.9Hh ...continued from previous page UTM (north)-WGS84 Zone: 32 Easting Northing Z Row data/Description Production source Statistical basis for normalized production: [m] [Months] 5 564,342 6,199,488 88.9 570715000000026669: 600 kW BONUS - Hundslund EmdEuropeEra5: 10.05E-55.97N-75m 196 6 564,728 6,199,104 81.0 570715000000026942: 600 kW BONUS - Hundslund EmdEuropeEra5: 10.05E-55.97N-75m 196 7 564,905 6,198,926 76.3 570715000000026959: 600 kW BONUS - Hundslund EmdEuropeEra5: 10.05E-55.97N-75m 196 8 564,606 6,204,046 81.2 570715000000058455: 900 kW NEG Micon - Torrild EmdEuropeEra5: 10.05E-55.97N-75m 196 9 564,601 6,204,295 85.0 570715000000058462: 900 kW NEG Micon - Torrild EmdEuropeEra5: 10.05E-55.97N-75m 196 10 564,599 6,204,545 79.3 570715000000058479: 900 kW NEG Micon - Torrild EmdEuropeEra5: 10.05E-55.97N-75m 196 11 563,818 6,196,778 60.0 571313113162149152: 2200 kW Vestas Wind Systems A/S - Trustrup EmdEuropeEra5: 10.05E-55.97N-75m 196 12 563,565 6,196,725 59.0 571313113162149176: 2200 kW Vestas Wind Systems A/S - Trustrup EmdEuropeEra5: 10.05E-55.97N-75m 196 13 567,416 6,214,281 60.7 570715000000023606: 750 kW NEG Micon - Blegind EmdEuropeEra5: 10.05E-55.97N-75m 196 14 567,328 6,214,403 59.2 570715000000023613: 750 kW NEG Micon - Blegind EmdEuropeEra5: 10.05E-55.97N-75m 196 15 567,240 6,214,525 60.7 570715000000023620: 750 kW NEG Micon - Blegind EmdEuropeEra5: 10.05E-55.97N-75m 196

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 3 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Production Analysis Calculation: 3x E82 E4 2.35MW 58.9HhWTG: All new WTGs, Air density varies with WTG position 1.239 kg/m³ - 1.248 kg/m³ Directional Analysis Sector 0 N 1 NNE 2 ENE 3 E 4 ESE 5 SSE 6 S 7 SSW 8 WSW 9 W 10 WNW 11 NNW Total Roughness based energy [MWh] 379.6 462.7 524.3 902.5 970.7 856.5 1,002.9 1,670.3 2,498.0 2,545.9 2,467.6 580.7 14,861.7 +Increase due to hills [MWh] 19.8 -0.4 -1.7 43.5 94.5 89.7 47.8 -1.1 -2.7 102.5 199.3 58.3 649.6 -Decrease due to wake losses [MWh] 0.3 0.0 0.9 256.7 2.1 0.0 0.8 17.0 3.6 607.1 26.3 9.6 924.5 Resulting energy [MWh] 399.0 462.3 521.7 689.3 1,063.1 946.2 1,049.8 1,652.1 2,491.8 2,041.3 2,640.6 629.4 14,586.8 Specific energy [kWh/m²] 921 Specific energy [kWh/kW] 2,069 Increase due to hills [%] 5.2 -0.1 -0.3 4.8 9.7 10.5 4.8 -0.1 -0.1 4.0 8.1 10.0 4.37 Decrease due to wake losses [%] 0.1 0.0 0.2 27.1 0.2 0.0 0.1 1.0 0.1 22.9 1.0 1.5 5.96 Utilization [%] 45.8 46.2 45.4 33.2 44.8 46.3 44.3 44.4 42.9 31.9 40.2 44.2 40.7 Operational [Hours/year] 335 348 423 650 666 647 658 905 1,177 1,178 1,215 484 8,687 Full Load Equivalent [Hours/year] 57 66 74 98 151 134 149 234 353 290 375 89 2,069

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 4 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Production Analysis Calculation: 3x E82 E4 2.35MW 58.9HhWTG: All existing WTGs, Air density varies with WTG position 1.239 kg/m³ - 1.248 kg/m³ Directional Analysis Sector 0 N 1 NNE 2 ENE 3 E 4 ESE 5 SSE 6 S 7 SSW 8 WSW 9 W 10 WNW 11 NNW Total Roughness based energy [MWh] 709.2 899.9 1,087.9 1,826.8 1,890.4 1,785.3 2,059.5 3,740.8 4,911.3 4,577.5 4,185.3 1,082.6 28,756.5 +Increase due to hills [MWh] 41.5 49.0 48.0 68.1 68.9 77.2 92.5 154.5 178.9 148.6 146.6 52.1 1,126.0 -Decrease due to wake losses [MWh] 34.6 9.9 64.7 158.3 38.0 98.5 74.4 14.0 215.9 313.0 54.0 54.2 1,129.4 Resulting energy [MWh] 716.1 939.0 1,071.2 1,736.7 1,921.3 1,764.1 2,077.7 3,881.3 4,874.3 4,413.1 4,277.9 1,080.4 28,753.1 Specific energy [kWh/m²] 849 Specific energy [kWh/kW] 2,435 Increase due to hills [%] 5.8 5.4 4.4 3.7 3.6 4.3 4.5 4.1 3.6 3.2 3.5 4.8 3.92 Decrease due to wake losses [%] 4.6 1.0 5.7 8.4 1.9 5.3 3.5 0.4 4.2 6.6 1.2 4.8 3.78 Utilization [%] 38.4 39.8 36.8 36.0 37.9 37.5 35.4 35.3 34.1 33.2 35.3 37.5 35.4 Operational [Hours/year] 302 322 390 580 575 562 584 828 1,075 1,048 1,047 428 7,741 Full Load Equivalent [Hours/year] 61 80 91 147 163 149 176 329 413 374 362 91 2,435

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 5 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Power Curve Analysis Calculation: 3x E82 E4 2.35MW 58.9HhWTG: 1 - ENERCON E-82 E4 2350 82.0 !O!, Hub height: 58.9 m Name: Level 2 - official - 2000kW s - 04/2016 Source: Enercon

Source/Date Created by Created Edited Stop wind speed Power control CT curve type Generator type Specific power [m/s] kW/m² 2016-04-13 EMD 2009-11-25 2016-06-28 25.0 Pitch User defined Variable 0.44 According to Enercon specification document D0388015-3_#_de_#_Betriebsmodus_E-82_E4___2350_kW_mit_TES.pdf

HP curve comparison - Note: For standard air density

Vmean [m/s] 5 6 7 8 9 10 HP value Pitch, variable speed (2013) [MWh] 2,780 4,477 6,269 7,981 9,510 10,792 ENERCON E-82 E4 2350 82.0 !O! Level 2 - official - 2000kW s - 04/2016 [MWh] 2,994 4,632 6,268 7,762 9,042 10,073 Check value [%] -7 -3 0 3 5 7 The table shows comparison between annual energy production calculated on basis of simplified "HP-curves" which assume that all WTGs performs quite similar - only specific power loading (kW/m^2) and single/dual speed or stall/pitch decides the calculated values. Productions are without wake losses. For further details, ask at the Danish Energy Agency for project report J.nr. 51171/00-0016 or see the windPRO manual. The method is refined in EMD report "20 Detailed Case Studies comparing Project Design Calculations and actual Energy Productions for Wind Energy Projects worldwide", jan 2003. Use the table to evaluate if the given power curve is reasonable - if the check value are lower than -5%, the power curve probably is too optimistic due to uncertainty in power curve measurement. Power curve Power, Efficiency and energy vs. wind speed Original data, Air density: 1.225 kg/m³ Data used in calculation, Air density: 1.239 kg/m³ New windPRO method (adjusted IEC Wind speed Power Cp Wind speed Ct curve method, improved to match turbine control) [m/s] [kW] [m/s] Wind speed Power Cp Interval Energy Acc.Energy Relative 0.0 0.0 0.00 0.0 0.00 [m/s] [kW] [m/s] [MWh] [MWh] [%] 1.0 0.0 0.00 1.0 0.00 1.0 0.0 0.00 0.50- 1.50 0.2 0.2 0.0 2.0 3.0 0.12 2.0 1.16 2.0 3.2 0.12 1.50- 2.50 4.3 4.5 0.1 3.0 25.0 0.29 3.0 0.97 3.0 25.6 0.29 2.50- 3.50 27.4 31.9 0.6 4.0 82.0 0.40 4.0 0.85 4.0 83.3 0.40 3.50- 4.50 92.5 124.4 2.5 5.0 174.0 0.43 5.0 0.88 5.0 176.6 0.43 4.50- 5.50 215.6 340.0 6.8 6.0 321.0 0.46 6.0 0.90 6.0 325.5 0.46 5.50- 6.50 392.9 732.9 14.7 7.0 532.0 0.48 7.0 0.91 7.0 539.1 0.48 6.50- 7.50 590.8 1,323.8 26.6 8.0 815.0 0.49 8.0 0.90 8.0 825.5 0.49 7.50- 8.50 754.0 2,077.8 41.8 9.0 1,180.0 0.50 9.0 0.84 9.0 1,191.8 0.50 8.50- 9.50 815.5 2,893.3 58.2 10.0 1,520.0 0.47 10.0 0.79 10.0 1,532.7 0.47 9.50-10.50 740.8 3,634.1 73.0 11.0 1,810.0 0.42 11.0 0.75 11.0 1,818.2 0.42 10.50-11.50 567.5 4,201.6 84.4 12.0 1,962.6 0.35 11.50-12.50 370.0 4,571.7 91.9 12.0 1,960.0 0.35 12.0 0.48 13.0 2,000.0 0.28 12.50-13.50 210.0 4,781.6 96.1 13.0 2,000.0 0.28 13.0 0.36 14.0 2,000.0 0.22 13.50-14.50 107.3 4,888.9 98.3 14.0 2,000.0 0.23 14.0 0.28 15.0 2,000.0 0.18 14.50-15.50 50.5 4,939.4 99.3 15.0 2,000.0 0.18 15.0 0.23 16.0 2,000.0 0.15 15.50-16.50 22.0 4,961.4 99.7 16.0 2,000.0 0.15 16.0 0.18 17.0 2,000.0 0.12 16.50-17.50 8.9 4,970.4 99.9 17.0 2,000.0 0.13 17.0 0.15 18.0 2,000.0 0.10 17.50-18.50 3.4 4,973.7 100.0 18.0 2,000.0 0.11 18.0 0.13 19.0 2,000.0 0.09 18.50-19.50 1.2 4,974.9 100.0 19.0 2,000.0 0.09 19.0 0.11 20.0 2,000.0 0.08 19.50-20.50 0.4 4,975.3 100.0 20.0 2,000.0 0.08 20.0 0.09 21.0 2,000.0 0.07 20.50-21.50 0.1 4,975.4 100.0 21.0 2,000.0 0.07 21.0 0.08 22.0 2,000.0 0.06 21.50-22.50 0.0 4,975.4 100.0 22.0 2,000.0 0.06 22.0 0.07 23.0 2,000.0 0.05 22.50-23.50 0.0 4,975.4 100.0 23.0 2,000.0 0.05 23.0 0.07 24.0 2,000.0 0.04 23.50-24.50 0.0 4,975.4 100.0 24.0 2,000.0 0.04 24.0 0.06 25.0 2,000.0 0.04 24.50-25.50 0.0 4,975.4 100.0 25.0 2,000.0 0.04 25.0 0.05

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 6 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Power Curve Analysis Calculation: 3x E82 E4 2.35MW 58.9HhWTG: 2 - ENERCON E-82 E4 2350 82.0 !O!, Hub height: 58.9 m Name: Level 0 - 2350kW - official - OM 0 - 2350kW - 06/2015 Source: Enercon

Source/Date Created by Created Edited Stop wind speed Power control CT curve type Generator type Specific power [m/s] kW/m² 2015-06-19 EMD 2009-11-25 2016-06-28 25.0 Pitch User defined Variable 0.44 According to Enercon specification document D0388022-1_#_de_#_Betriebsmodus_E-82_E4___2350_kW.pdf

HP curve comparison - Note: For standard air density

Vmean [m/s] 5 6 7 8 9 10 HP value Pitch, variable speed (2013) [MWh] 2,780 4,477 6,269 7,981 9,510 10,792 ENERCON E-82 E4 2350 82.0 !O! Level 0 - 2350kW - official - OM 0 - 2350kW - 06/2015 [MWh] 3,050 4,801 6,614 8,321 9,823 11,062 Check value [%] -9 -7 -5 -4 -3 -2 The table shows comparison between annual energy production calculated on basis of simplified "HP-curves" which assume that all WTGs performs quite similar - only specific power loading (kW/m^2) and single/dual speed or stall/pitch decides the calculated values. Productions are without wake losses. For further details, ask at the Danish Energy Agency for project report J.nr. 51171/00-0016 or see the windPRO manual. The method is refined in EMD report "20 Detailed Case Studies comparing Project Design Calculations and actual Energy Productions for Wind Energy Projects worldwide", jan 2003. Use the table to evaluate if the given power curve is reasonable - if the check value are lower than -5%, the power curve probably is too optimistic due to uncertainty in power curve measurement. Power curve Power, Efficiency and energy vs. wind speed Original data, Air density: 1.225 kg/m³ Data used in calculation, Air density: 1.239 kg/m³ New windPRO method (adjusted IEC Wind speed Power Cp Wind speed Ct curve method, improved to match turbine control) [m/s] [kW] [m/s] Wind speed Power Cp Interval Energy Acc.Energy Relative 0.0 0.0 0.00 0.0 0.00 [m/s] [kW] [m/s] [MWh] [MWh] [%] 1.0 0.0 0.00 1.0 0.00 1.0 0.0 0.00 0.50- 1.50 0.2 0.2 0.0 2.0 3.0 0.12 2.0 1.16 2.0 3.2 0.12 1.50- 2.50 4.1 4.3 0.1 3.0 25.0 0.29 3.0 0.97 3.0 25.6 0.29 2.50- 3.50 26.0 30.3 0.6 4.0 82.0 0.40 4.0 0.85 4.0 83.3 0.40 3.50- 4.50 87.8 118.2 2.4 5.0 174.0 0.43 5.0 0.88 5.0 176.6 0.43 4.50- 5.50 204.6 322.8 6.6 6.0 321.0 0.46 6.0 0.90 6.0 325.5 0.46 5.50- 6.50 372.8 695.5 14.3 7.0 532.0 0.48 7.0 0.91 7.0 539.1 0.48 6.50- 7.50 560.3 1,255.9 25.8 8.0 815.0 0.49 8.0 0.90 8.0 825.5 0.49 7.50- 8.50 714.9 1,970.8 40.5 9.0 1,180.0 0.50 9.0 0.84 9.0 1,193.9 0.50 8.50- 9.50 782.4 2,753.2 56.6 10.0 1,580.0 0.49 10.0 0.79 10.0 1,593.6 0.49 9.50-10.50 725.4 3,478.6 71.5 11.0 1,890.0 0.44 11.0 0.75 11.0 1,901.6 0.44 10.50-11.50 565.7 4,044.3 83.1 12.0 2,109.9 0.37 11.50-12.50 378.9 4,423.3 90.9 12.0 2,100.0 0.38 12.0 0.63 13.0 2,257.2 0.31 12.50-13.50 224.9 4,648.1 95.5 13.0 2,250.0 0.32 13.0 0.44 14.0 2,350.0 0.26 13.50-14.50 119.9 4,768.0 97.9 14.0 2,350.0 0.26 14.0 0.34 15.0 2,350.0 0.21 14.50-15.50 57.7 4,825.8 99.1 15.0 2,350.0 0.22 15.0 0.27 16.0 2,350.0 0.18 15.50-16.50 25.6 4,851.4 99.7 16.0 2,350.0 0.18 16.0 0.22 17.0 2,350.0 0.15 16.50-17.50 10.6 4,862.0 99.9 17.0 2,350.0 0.15 17.0 0.18 18.0 2,350.0 0.12 17.50-18.50 4.1 4,866.1 100.0 18.0 2,350.0 0.12 18.0 0.15 19.0 2,350.0 0.10 18.50-19.50 1.5 4,867.5 100.0 19.0 2,350.0 0.11 19.0 0.13 20.0 2,350.0 0.09 19.50-20.50 0.5 4,868.0 100.0 20.0 2,350.0 0.09 20.0 0.11 21.0 2,350.0 0.08 20.50-21.50 0.2 4,868.2 100.0 21.0 2,350.0 0.08 21.0 0.10 22.0 2,350.0 0.07 21.50-22.50 0.0 4,868.2 100.0 22.0 2,350.0 0.07 22.0 0.09 23.0 2,350.0 0.06 22.50-23.50 0.0 4,868.2 100.0 23.0 2,350.0 0.06 23.0 0.08 24.0 2,350.0 0.05 23.50-24.50 0.0 4,868.3 100.0 24.0 2,350.0 0.05 24.0 0.07 25.0 2,350.0 0.05 24.50-25.50 0.0 4,868.3 100.0 25.0 2,350.0 0.05 25.0 0.06

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 7 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Power Curve Analysis Calculation: 3x E82 E4 2.35MW 58.9HhWTG: 3 - ENERCON E-82 E4 2350 82.0 !O!, Hub height: 58.9 m Name: Level 0 - 2350kW - official - OM 0 - 2350kW - 06/2015 Source: Enercon

Source/Date Created by Created Edited Stop wind speed Power control CT curve type Generator type Specific power [m/s] kW/m² 2015-06-19 EMD 2009-11-25 2016-06-28 25.0 Pitch User defined Variable 0.44 According to Enercon specification document D0388022-1_#_de_#_Betriebsmodus_E-82_E4___2350_kW.pdf

HP curve comparison - Note: For standard air density

Vmean [m/s] 5 6 7 8 9 10 HP value Pitch, variable speed (2013) [MWh] 2,780 4,477 6,269 7,981 9,510 10,792 ENERCON E-82 E4 2350 82.0 !O! Level 0 - 2350kW - official - OM 0 - 2350kW - 06/2015 [MWh] 3,050 4,801 6,614 8,321 9,823 11,062 Check value [%] -9 -7 -5 -4 -3 -2 The table shows comparison between annual energy production calculated on basis of simplified "HP-curves" which assume that all WTGs performs quite similar - only specific power loading (kW/m^2) and single/dual speed or stall/pitch decides the calculated values. Productions are without wake losses. For further details, ask at the Danish Energy Agency for project report J.nr. 51171/00-0016 or see the windPRO manual. The method is refined in EMD report "20 Detailed Case Studies comparing Project Design Calculations and actual Energy Productions for Wind Energy Projects worldwide", jan 2003. Use the table to evaluate if the given power curve is reasonable - if the check value are lower than -5%, the power curve probably is too optimistic due to uncertainty in power curve measurement. Power curve Power, Efficiency and energy vs. wind speed Original data, Air density: 1.225 kg/m³ Data used in calculation, Air density: 1.239 kg/m³ New windPRO method (adjusted IEC Wind speed Power Cp Wind speed Ct curve method, improved to match turbine control) [m/s] [kW] [m/s] Wind speed Power Cp Interval Energy Acc.Energy Relative 0.0 0.0 0.00 0.0 0.00 [m/s] [kW] [m/s] [MWh] [MWh] [%] 1.0 0.0 0.00 1.0 0.00 1.0 0.0 0.00 0.50- 1.50 0.2 0.2 0.0 2.0 3.0 0.12 2.0 1.16 2.0 3.2 0.12 1.50- 2.50 4.3 4.5 0.1 3.0 25.0 0.29 3.0 0.97 3.0 25.6 0.29 2.50- 3.50 27.0 31.5 0.7 4.0 82.0 0.40 4.0 0.85 4.0 83.4 0.40 3.50- 4.50 90.9 122.4 2.6 5.0 174.0 0.43 5.0 0.88 5.0 176.7 0.43 4.50- 5.50 210.3 332.7 7.0 6.0 321.0 0.46 6.0 0.90 6.0 325.7 0.46 5.50- 6.50 380.3 713.0 15.0 7.0 532.0 0.48 7.0 0.91 7.0 539.4 0.48 6.50- 7.50 566.2 1,279.3 27.0 8.0 815.0 0.49 8.0 0.90 8.0 826.0 0.49 7.50- 8.50 714.1 1,993.3 42.0 9.0 1,180.0 0.50 9.0 0.84 9.0 1,194.5 0.50 8.50- 9.50 770.6 2,763.9 58.3 10.0 1,580.0 0.49 10.0 0.79 10.0 1,594.1 0.49 9.50-10.50 702.6 3,466.5 73.1 11.0 1,890.0 0.44 11.0 0.75 11.0 1,902.0 0.44 10.50-11.50 537.0 4,003.5 84.4 12.0 2,110.2 0.37 11.50-12.50 351.2 4,354.7 91.8 12.0 2,100.0 0.38 12.0 0.63 13.0 2,257.5 0.31 12.50-13.50 202.6 4,557.3 96.1 13.0 2,250.0 0.32 13.0 0.44 14.0 2,350.0 0.26 13.50-14.50 104.5 4,661.7 98.3 14.0 2,350.0 0.26 14.0 0.34 15.0 2,350.0 0.21 14.50-15.50 48.4 4,710.1 99.3 15.0 2,350.0 0.22 15.0 0.27 16.0 2,350.0 0.18 15.50-16.50 20.5 4,730.6 99.7 16.0 2,350.0 0.18 16.0 0.22 17.0 2,350.0 0.15 16.50-17.50 8.1 4,738.7 99.9 17.0 2,350.0 0.15 17.0 0.18 18.0 2,350.0 0.12 17.50-18.50 3.0 4,741.7 100.0 18.0 2,350.0 0.12 18.0 0.15 19.0 2,350.0 0.10 18.50-19.50 1.0 4,742.7 100.0 19.0 2,350.0 0.11 19.0 0.13 20.0 2,350.0 0.09 19.50-20.50 0.3 4,743.0 100.0 20.0 2,350.0 0.09 20.0 0.11 21.0 2,350.0 0.08 20.50-21.50 0.1 4,743.1 100.0 21.0 2,350.0 0.08 21.0 0.10 22.0 2,350.0 0.07 21.50-22.50 0.0 4,743.1 100.0 22.0 2,350.0 0.07 22.0 0.09 23.0 2,350.0 0.06 22.50-23.50 0.0 4,743.1 100.0 23.0 2,350.0 0.06 23.0 0.08 24.0 2,350.0 0.05 23.50-24.50 0.0 4,743.1 100.0 24.0 2,350.0 0.05 24.0 0.07 25.0 2,350.0 0.05 24.50-25.50 0.0 4,743.1 100.0 25.0 2,350.0 0.05 25.0 0.06

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 8 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Terrain Calculation: 3x E82 E4 2.35MW 58.9HhSite Data: A - Site data: WAsP EMD Europe

Obstacles:

0 Obstacles used

Roughness:

Terrain data files used in calculation: C:\Users\madalina\Documents\01_Projects\19179\WindPRO\ROUGHNESSLINE_Torrild_0.wpo Min X: 536,095, Max X: 596,002, Min Y: 6,171,652, Max Y: 6,233,515, Width: 59,907 m, Height: 61,863 m

Orography:

Terrain data files used in calculation: C:\Users\madalina\Documents\01_Projects\19179\WindPRO\CONTOURLINE_ONLINEDATA_0.wpo Min X: 551,133, Max X: 580,117, Min Y: 6,187,333, Max Y: 6,218,043, Width: 28,984 m, Height: 30,711 m

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 9 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Wind Data Analysis Calculation: 3x E82 E4 2.35MW 58.9HhWind data: A - Site data: WAsP EMD Europe; Hub height: 50.0 Site coordinates Weibull Data UTM (north)-WGS84 Zone: 32 Current site Reference: Roughness class 1 East: 565,493 North: 6,203,779 Sector A- Wind k- Frequency A- k- Frequency Wind statistics parameter speed parameter parameter parameter DK EmdEuropeEra5_N55.974922_E010.049438 (2) - J Scale 58.90 m.wws [m/s] [m/s] [%] [m/s] [%] 0 N 5.57 4.93 2.240 3.8 6.17 2.264 3.9 1 NNE 6.06 5.37 2.436 4.2 6.72 2.498 4.1 2 ENE 6.15 5.45 2.252 5.2 6.44 2.244 5.0 3 E 6.35 5.63 2.447 7.8 6.82 2.447 7.6 4 ESE 6.29 5.58 2.443 7.4 6.84 2.436 7.4 5 SSE 6.08 5.40 2.607 7.0 6.68 2.607 7.2 6 S 6.24 5.53 2.334 7.3 6.95 2.338 7.5 7 SSW 7.31 6.49 2.643 10.7 7.88 2.682 10.5 8 WSW 7.64 6.78 2.584 14.4 8.19 2.635 13.7 9 W 7.54 6.69 2.486 13.9 8.15 2.553 13.5 10 WNW 7.18 6.36 2.369 13.3 7.82 2.400 13.6 11 NNW 5.58 4.94 2.170 5.2 6.49 2.189 6.0 All 6.80 6.02 2.365 100.0 7.37 2.406 100.0

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 10 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Park power curve Calculation: 3x E82 E4 2.35MW 58.9Hh Power Wind speed Free WTGs Park WTGs N NNE ENE E ESE SSE S SSW WSW W WNW NNW [m/s] [kW] [kW] [kW] [kW] [kW] [kW] [kW] [kW] [kW] [kW] [kW] [kW] [kW] [kW] 0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1.5 5 4 5 5 5 4 5 5 5 5 5 4 5 5 2.5 43 38 43 43 42 23 42 43 43 42 42 23 42 43 3.5 163 148 162 163 162 99 162 163 162 159 162 99 158 157 4.5 388 360 388 388 388 264 387 388 388 382 388 264 380 378 5.5 751 697 751 751 750 516 749 751 750 740 750 516 739 737 6.5 1,294 1,200 1,293 1,294 1,291 884 1,291 1,294 1,293 1,275 1,291 883 1,273 1,269 7.5 2,044 1,896 2,042 2,044 2,039 1,400 2,038 2,044 2,041 2,014 2,039 1,400 2,008 2,001 8.5 3,028 2,819 3,025 3,028 3,022 2,112 3,020 3,028 3,025 2,987 3,022 2,112 2,983 2,975 9.5 4,155 3,900 4,153 4,155 4,149 3,038 4,148 4,155 4,153 4,115 4,149 3,006 4,109 4,105 10.5 5,181 4,923 5,178 5,181 5,175 4,048 5,175 5,181 5,178 5,154 5,175 3,994 5,147 5,142 11.5 5,910 5,724 5,909 5,910 5,907 5,045 5,907 5,910 5,909 5,896 5,907 5,063 5,889 5,890 12.5 6,354 6,254 6,353 6,354 6,353 5,902 6,352 6,354 6,354 6,346 6,352 5,894 6,341 6,345 13.5 6,615 6,574 6,615 6,615 6,615 6,479 6,615 6,615 6,615 6,611 6,614 6,399 6,607 6,612 14.5 6,700 6,693 6,700 6,700 6,700 6,686 6,700 6,700 6,700 6,700 6,700 6,656 6,700 6,700 15.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 16.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 17.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 18.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 19.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 20.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 21.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 22.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 23.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 24.5 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 6,700 25.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 26.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 27.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 28.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 29.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Description: The park power curve is similar to a WTG power curve, meaning that when a given wind speed appears in front of the park with same speed in the entire wind farm area (before influence from the park), the output from the park can be found in the park power curve. Another way to say this: The park power curve includes wake losses, but do NOT include terrain given variations in the wind speed over the park area. Measuring a park power curve is not as simple as measuring a WTG power curve due to the fact that the park power curve depends on the wind direction and that the same wind speed normally will not appear for the entire park area at the same time (only in very flat non-complex terrain). The idea with this version of the park power curve is not to use it for validation based on measurements. This would require at least 2 measurement masts at two sides of the park, unless only a few direction sectors should be tested, AND non complex terrain (normally only useable off shore). Another park power curve version for complex terrain is available in windPRO.

The park power curve can be used for: 1. Forecast systems, based on more rough (approximated) wind data, the park power curve would be an efficient way to make the connection from wind speed (and direction) to power. 2. Construction of duration curves, telling how often a given power output will appear, the park power curve can be used together with the average wind distribution for the Wind farm area in hub height. The average wind distribution can eventually be obtained based on the Weibull parameters for each WTG position. These are found at print menu: >Result to file< in the >Park result< which can be saved to file or copied to clipboard and pasted in Excel. 3. Calculation of wind energy index based on the PARK production (see below). 4. Estimation of the expected PARK production for an existing wind farm based on wind measurements at minimum 2 measurement masts at two sides of wind farm. The masts must be used for obtaining the free wind speed. The free wind speed is used in the simulation of expected energy production with the PARK power curve. This procedure will only work suitable in non complex terrains. For complex terrain another park power curve calculation is available in windPRO (PPV-model).

Note: From the >Result to file< the >Wind Speeds Inside Wind farm< is also available. These can (e.g. via Excel) be used for extracting the wake induced reductions in measured wind speed.

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 11 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - WTG distances Calculation: 3x E82 E4 2.35MW 58.9Hh WTG distances Z Nearest WTG Z Horizontal Distance in Distance in distance rotor rotor diameters diameters [m] [m] [m] (max) (min) 1 80.0 2 79.8 353 4.3 4.3 2 79.8 3 75.5 353 4.3 4.3 3 75.5 2 79.8 353 4.3 4.3 4 13.3 3 75.5 11,741 249.8 143.2 5 88.9 6 81.0 544 12.4 12.4 6 81.0 7 76.3 252 5.7 5.7 7 76.3 6 81.0 252 5.7 5.7 8 81.2 9 85.0 249 4.8 4.8 9 85.0 8 81.2 249 4.8 4.8 10 79.3 9 85.0 250 4.8 4.8 11 60.0 12 59.0 258 2.6 2.6 12 59.0 11 60.0 258 2.6 2.6 13 60.7 14 59.2 151 3.1 3.1 14 59.2 15 60.7 150 3.1 3.1 15 60.7 14 59.2 150 3.1 3.1 (C) OpenStreetMap contributors, Data OpenStreetMap and contributors, ODbL Min 13.3 59.0 150 2.6 2.6 Scale 1:400,000 Max 88.9 85.0 11,741 249.8 143.2 New WTG Existing WTG

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 12 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Wind statistics info Calculation: 3x E82 E4 2.35MW 58.9Hh Main data for wind statistic File C:\Users\madalina\Documents\01_Projects\19179\WindPRO\DK EmdEuropeEra5_N55.974922_E010.049438 (2) - J Scale 58.90 m.wws Name EmdEuropeEra5_N55.974922_E010.049438 (2) - J Scale 58.90 m Country Denmark Source USER Mast coordinates UTM (north)-WGS84 Zone: 32 East: 565,492 North: 6,203,786 Created 2019-12-18 Edited 2020-01-08 Sectors 12 WAsP version WAsP 11 Version 11.06.0028 Coordinate system UTM (north)-WGS84 Zone: 32 Displacement height None

Additional info for wind statistic Source data EmdEuropeEra5_N55.974922_E010.049438 (2) Data from 1999-09-01 Data to 2019-08-31 Measurement length 240.0 Months Recovery rate 100.0 % Effective measurement length 240.0 Months

Note To get the most correct calculation results, wind statistics shall be calculated with the SAME model and model parameters, as currently chosen in calculation. For WAsP versions before 10.0, the model is unchanged, but thereafter more model changes affecting the wind statistic is seen. Likewise WAsP CFD should always use WAsP CFD calculated wind statistics.

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 13 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Displacement height Calculation: 3x E82 E4 2.35MW 58.9Hh

Sector-wise from calculator: Ustund Forest description is based on area object with heights

Calculation uses: C:\Users\madalina\Documents\01_Projects\19179\WindPRO\REGIONS_Torrild_2.w2r

WTG displacement height set to 1.00 of forest height. WTG displacement height are adjusted to a factor of 50.0 of forest height in up wind direction. WTG displacement height are adjusted to a factor of 25.0 of forest height in down wind direction.

Sector wise displacement heights for WTGs Valid Manufact. Type-generator Power, Rotor Hub DH (0) DH (1) DH (2) DH (3) DH (4) DH (5) DH (6) DH (7) DH (8) DH (9) DH DH rated diameter height (10) (11) [kW] [m] [m] [m] [m] [m] [m] [m] [m] [m] [m] [m] [m] [m] [m] 1 Yes ENERCON E-82 E4-2,350 2,350 82.0 58.9 0.00 0.00 0.00 0.00 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2 Yes ENERCON E-82 E4-2,350 2,350 82.0 58.9 0.00 0.00 0.00 0.09 0.00 1.77 3.04 0.00 0.00 0.00 0.00 0.00 3 Yes ENERCON E-82 E4-2,350 2,350 82.0 58.9 0.00 0.00 1.44 2.63 6.84 0.00 0.00 1.97 0.47 0.00 3.68 0.00 4 No VESTAS V47-660/200 660 47.0 50.0 7.22 9.80 9.91 9.92 9.92 9.86 7.56 9.60 9.82 9.85 9.83 9.72 5 No BONUS MK IV-600/120 600 44.0 45.0 0.73 0.57 0.00 0.69 2.08 0.00 0.00 0.00 0.23 4.46 3.04 0.05 6 No BONUS MK IV-600/120 600 44.0 45.0 2.76 9.21 1.83 0.00 0.00 0.09 3.29 8.42 1.67 0.00 0.00 0.00 7 No BONUS MK IV-600/120 600 44.0 45.0 0.00 0.00 0.00 0.00 0.00 2.61 3.54 3.94 0.47 0.00 0.00 1.41 8 No NEG MICON NM52/900-900/200 900 52.0 49.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 9 No NEG MICON NM52/900-900/200 900 52.0 49.0 2.03 1.34 0.35 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 10 No NEG MICON NM52/900-900/200 900 52.0 49.0 7.94 7.57 7.53 2.04 0.00 1.73 5.88 5.13 5.06 0.58 0.00 2.37 11 No VESTAS V100-2,200 2,200 100.0 75.0 0.00 0.00 0.05 2.04 0.04 0.00 0.00 0.00 0.00 2.58 4.25 1.29 12 No VESTAS V100-2,200 2,200 100.0 75.0 7.25 0.00 0.00 0.16 0.00 1.26 5.50 0.00 0.29 0.84 1.78 4.57 13 No NEG MICON NM48/750-750/200 750 48.2 43.6 0.00 0.00 0.00 0.65 0.15 0.00 0.00 0.00 0.00 1.75 0.00 0.00 14 No NEG MICON NM48/750-750/200 750 48.2 45.0 0.26 0.46 0.00 0.00 0.00 0.00 0.00 1.23 0.00 0.40 2.35 0.00 15 No NEG MICON NM48/750-750/200 750 48.2 45.0 1.70 0.00 0.00 0.00 0.09 0.00 0.89 0.00 1.07 2.81 4.05 0.00

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 14 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material EMD International A/S caused by errors or omissions in the delivered data. EMD cannot be held liable for erroneous results Niels Jernes Vej 101 caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø +45 9635 4444 For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of Madalina / [email protected] the form of action, shall be limited to the total amount paid to EMD for the services provided as part Calculated: of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the expense of the Client. 2020-01-10 09:14/3.4.241 PARK - Map Calculation: 3x E82 E4 2.35MW 58.9Hh

(C) OpenStreetMap contributors, Data OpenStreetMap and contributors, ODbL

0 2.5 5 7.5 10 km Map: EMD OpenStreetMap , Print scale 1:125,000, Map center UTM (north)-WGS84 Zone: 32 East: 570,230 North: 6,205,625 New WTG Existing WTG Area object (Heights a.g.l. for e.g. Forest (ORA tool) or ZVI obstructions): REGIONS_Torrild_2.w2r (2) windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:22 / 15 windPRO

26/26

Appendix B – Net AEP calculation

200110_19179_A_MC_0_ts2 EMD International A/S Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material caused by EMD International A/S errors or omissions in the delivered data. EMD cannot Niels Jernes Vej 101 be held liable for erroneous results caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of the form of +45 9635 4444 action, shall be limited to the total amount paid to EMD Madalina / [email protected] for the services provided as part of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the Calculated: expense of the Client. 2020-01-10 09:16/3.4.241 Loss&Uncertainty - Main result Calculation: 3 x E82 E4 2.35MW 58.9mHH Main data for PARK PARK calculation 3.4.241: 3x E82 E4 2.35MW 58.9Hh Count 3 Rated power 7.1 MW Mean wind speed 6.4 m/s at hub height Sensitivity 2.4 %AEP / %Mean Wind Speed Expected lifetime 20 Years

RESULTS P50 P84 P90 NET AEP [MWh/y] 14,098 12,524 12,070 Capacity factor [%] 22.8 20.3 19.5 Full load hours [h/y] 2,000 1,777 1,712

(C) OpenStreetMap contributors, Data OpenStreetMap and contributors, ODbL Scale: 10,000 Result details

P50 Uncertainty GROSS AEP *) 15,511 MWh/y 11.0 % Bias correction 310 MWh/y 2.0 % 0.4 % Loss correction -1,723 MWh/y -10.9 % 1.9 % Wake loss -6.0 % Other losses -5.2 % NET AEP 14,098 MWh/y 11.2 %

*) Calculated Annual Energy Production before any bias or loss corrections Assumptions: Uncertainty and percentiles (PXX values) are calculated for the expected lifetime windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:16 / 1 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material caused by EMD International A/S errors or omissions in the delivered data. EMD cannot Niels Jernes Vej 101 be held liable for erroneous results caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of the form of +45 9635 4444 action, shall be limited to the total amount paid to EMD Madalina / [email protected] for the services provided as part of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the Calculated: expense of the Client. 2020-01-10 09:16/3.4.241 Loss&Uncertainty - Assumptions and results Calculation: 3 x E82 E4 2.35MW 58.9mHH ASSUMPTIONS

BIAS Method *) Correction, Correction, Std dev**) Comment wind speed AEP [%] [%] [%] Other bias Estimate 0.8 2.0 20.0 1% for electrical losses and 1 % for availability losses BIAS, total 2.0 0.4

LOSS Method *) Loss Loss Std dev**) Comment [%] [MWh/y] [%] 1. Wake effects Wake effects, all WTGs Calculation 6.0 943 20.0 2. Availability Turbine availability Estimate 3.0 475 50.0 Other availability Estimate 0.6 95 25.0 force major 3. Turbine performance No input 4. Electrical Electrical losses Estimate 1.0 158 25.0 5. Environmental No input 6. Curtailment Power purchase agreement curtailment Estimate 0.7 115 25.0 power limited to 6.1MW 7. Other No input LOSS, total 10.9 1,723 1.9

UNCERTAINTY Method *) Std dev, Std dev, Comment wind speed AEP [%] [%] A. Wind data Wind measurement/Wind data Estimate 3.0 7.1 Long term correction Estimate 3.0 Year-to-year variability Estimate 3.1 7.4 Future climate Estimate 1.5 3.6 Reference WTGs Other wind related B. Wind model Vertical extrapolation Estimate 1.0 2.4 Horizontal extrapolation Estimate 1.3 3.1 Uncertainty of Terrain data Other wind model related Estimate 2.0 4.8 tree growth C. Power conversion Power curve uncertainty Estimate 2.9 Metering uncertainty Estimate 0.3 Site-specific impacts on power curve Differing technical operating behavior Other AEP related uncertainties D. BIAS, total uncertainty 0.4 E. LOSS, total uncertainty 1.9 UNCERTAINTY, total (1y average) 13.3 UNCERTAINTY, total (20y average) 11.2

VARIABILITY Years Variability Total (std dev) std dev [%] [%] 1 7.39 13.3 5 3.30 11.6 10 2.34 11.3 20 1.65 11.2

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:16 / 2 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material caused by EMD International A/S errors or omissions in the delivered data. EMD cannot Niels Jernes Vej 101 be held liable for erroneous results caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of the form of +45 9635 4444 action, shall be limited to the total amount paid to EMD Madalina / [email protected] for the services provided as part of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the Calculated: expense of the Client. 2020-01-10 09:16/3.4.241 Loss&Uncertainty - Assumptions and results Calculation: 3 x E82 E4 2.35MW 58.9mHH RESULTS

AEP versus exceedance level / time horizon PXX 1 y 5 y 10 y 20 y [%] [MWh/y] [MWh/y] [MWh/y] [MWh/y] 50 14,098 14,098 14,098 14,098 75 12,830 12,997 13,019 13,031 84 12,229 12,474 12,507 12,524 90 11,689 12,005 12,048 12,070 95 11,006 11,412 11,467 11,495

*) Calculation means that a calculation method available in the windPRO software is used. This still typically involve a user judgement and user data where the quality of those decides the accuracy. If calculation method is used, the values will often be different from turbine to turbine, here the average is shown, but at page "WTG results" the individual turbine results are shown. **) For totals the std dev refers to the full AEP, otherwise std dev refers to the bias or loss component which is a fraction of the total AEP.

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:16 / 3 windPRO Project: Description: Licensed user: Torrild EMD International A/S (EMD) does not warrant, guarantee or make any representations regarding the delivered consultancy material caused by EMD International A/S errors or omissions in the delivered data. EMD cannot Niels Jernes Vej 101 be held liable for erroneous results caused by inaccuracy, limitations or malfunctioning of models or software used. DK-9220 Aalborg Ø For any claim whatsoever related to the subject matter of this consultancy job, the liability of EMD for actual damages, regardless of the form of +45 9635 4444 action, shall be limited to the total amount paid to EMD Madalina / [email protected] for the services provided as part of this consultancy job. Separate insurance cover for extended liability can be provided upon request, but at the Calculated: expense of the Client. 2020-01-10 09:16/3.4.241 Loss&Uncertainty - WTG results Calculation: 3 x E82 E4 2.35MW 58.9mHH Main data for PARK PARK calculation 3.4.241: 3x E82 E4 2.35MW 58.9Hh Count 3 Rated power 7.1 MW Mean wind speed 6.4 m/s at hub height Sensitivity 2.4 %AEP / %Mean Wind Speed Expected lifetime 20 Years

(C) OpenStreetMap contributors, Data OpenStreetMap and contributors, ODbL Scale: 10,000 Expected AEP per WTG including bias, loss and uncertainty evaluation 20 years averaging Description User label Calculated GROSS*) Bias Loss Unc. P50 P84 P90 [MWh/y] [%] [%] [%] [MWh/y] [MWh/y] [MWh/y] 1 ENERCON E-82 E4 2350 82.0 !O! hub: 58.9 m (TOT: 99.9 m) (2) 1 5,127.0 2.0 8.0 10.9 4,808.9 4,289.6 4,139.7 2 ENERCON E-82 E4 2350 82.0 !O! hub: 58.9 m (TOT: 99.9 m) (3) 2 5,283.4 2.0 12.7 11.4 4,705.3 4,173.9 4,020.5 3 ENERCON E-82 E4 2350 82.0 !O! hub: 58.9 m (TOT: 99.9 m) (4) 3 5,100.8 2.0 11.9 11.5 4,584.3 4,060.6 3,909.3 PARK 15,511.3 2.0 10.9 11.2 14,098.5 12,524.4 12,070.0

windPRO 3.4.241 Beta by EMD International A/S, Tel. +45 96 35 44 44, www.emd.dk, [email protected] 2020-01-10 09:16 / 4 windPRO