wind measurement strategies to optimize lidar … · wind measurement strategies to optimize lidar...
TRANSCRIPT
Wind Measurement Strategies
to Optimize Lidar Return on Investment
Matthieu Boquet1, Karin Görner
2, Kai Mönnich
2
1LEOSPHERE SAS, 14-16 avenue Jean Rostand, 91400 Orsay, France, [email protected]
2DEWI GmbH, Kasinoplatz 3, 26122 Oldenburg, Germany, [email protected]
Abstract Understanding the wind resource at a prospective project site has long been considered a critical step in the wind farm development process, and therefore wind resource experts have become more and more sophisticated in performing the assessment of the wind resource. The data collected from a wind resource assessment program, and the accuracy of that data, drives the success of the wind farm project. In the context of the constant aim to reduce project uncertainties through the design of their wind resource estimate campaigns, consultants make use of new measurement technologies and methods of analyzing data. Though the combination of met masts and lidars is one approach that is gaining traction, a remaining question is which combination strategy must be applied to reach greatest uncertainties reduction at reasonable operational costs.
Methodology In this paper, DEWI and LEOSPHERE propose to study various wind measurement strategies on a representative wind farm site. Several measurement system combinations are proposed, including met masts of different heights and lidar devices, located at one or several locations for varying duration and seasonal periods. The resulting uncertainties on the annual energy yield estimation are calculated and compared. Based on a wind farm business case, the results are also shown on a financial level, taking into account the operational costs, leveraging effect, equity investment and Internal Rate of Return from the developer perspective. The best measurement strategies are highlighted according to a higher Return on Investment of the costs involved in the assessment of the wind resource.
Site description A theoretical, but realistic wind farm layout with 41 wind turbine positions embedded in a medium-complex terrain with an elevation ranging from 0 – 90 m has been used for the study. Mast and lidar positions were assumed in about 4 km distances to gain an optimum coverage of the wind farm area. Lidar measurements, when placed away from the mast measurement, are useful to gain additional information about the spatial distribution of the wind resource and help to reduce model calculation uncertainty with respect to horizontal extrapolation. A lidar placed close to the mast can reduce the uncertainty in vertical extrapolation, if the mast height is lower than the planned hub height. When considering site-specific conditions, a well-planned combined measurement strategy can therefore lower the annual energy yield estimation within a study.
Figure 1 Site layout with turbines, mast and lidar positions
Measurement strategies Different measurement strategies comprising mast and lidar system(s) in several combinations as defined in the following table have been studied for a wind farm with planned hub height of 100 m. Positions of masts are fixed at one location for the complete 1 year of assessment. For cost-effectiveness reasons lidar systems are usually measuring only a few months at one location and site. But their mobility could be used to move them on site several times during the 1 year period. Two different strategies of moving a lidar system on-site have been considered and following lidar measurement periods have been assumed in the study:
• Measuring fixed at one location during 3 successive months • Measuring at one location 1 month in each season (in sum 4 months)
In the second case the lidar measurement is more representative regarding wind speed distribution and wind profile compared to a 1 year measurement period (assuming Mid-European regions) and therefore more suitable for e.g. time series correlation.
Table 1
The strategy M080Lc1Laf2 means having an 80m met mast with a lidar being positioned close to the mast for one period, and then positioned away at two locations during successive periods. The strategy M100Las3 means having a 100m met mast with a lidar being located away from the mast at 3 different locations. The lidar is moved within the seasons to probe every location once in each season.
Energy Yield Assessment Uncertainties Based on the theoretical layout the uncertainty of an energy yield calculation (with WASP) has been determined assuming the following:
• Measurement uncertainty mast: 2.0% (high quality, IEC and IEA conform) • Measurement uncertainty lidar: 2.0% (WINDCUBE®v2 lidar, following installation
best practices) • Uncertainty in long-term correlation of mast: 3.9% • Average sensitivity factor (dE/dv): 1.85
Main focus is given on the uncertainty reduction of horizontal and vertical extrapolation assuming additional available wind data information of same quality as mast measurement data for energy yield modeling. It has to be noted that the remote sensing device uncertainty can vary among the different types of sensor and according to the site complexity. The device used here is a WINDCUBE®v2 lidar which uncertainty has universally and independently been assessed by several wind experts ([3], [4], [7]), and is here taken same as well calibrated cups. Furthermore, uncertainties resulting from time series correlation with mast data or seasonal correction of a short-term measured wind profile during 3 successive months are site-specific and have been assumed here as low. This study is aiming at gaining sensitivity on the uncertainty reduction by using a lidar, and accumulation of on-site measurement experiences will help refine these uncertainties reductions. Different uncertainty values were determined for the vertical and horizontal extrapolation uncertainty for each measurement strategy. The results are summarized in the table below:
Table 2
Costs of the Measurement Strategies Operational costs for conducting the different strategies are here studied with the following mean market values in 2010.
Table 3
Further assumptions:
• Lidar price includes options like safety and enhanced communication skills etc…
• The “on site move” cost is used when the lidar is moved from one location to another
within the same project site. It mainly considers the working time on-site. If the lidar is necessary less than a year on the same site, it is sent to another project to get a full year use of the instrument. The installation and dismantlement costs are then applied.
• Amortization of masts and lidar is 3 years
Finally the total costs of operation (TCO) of the strategies are summarized below:
str
ate
gy
M100
M0
80
M0
60
M0
80-L
c1
M0
60-L
c1
M0
80-L
c1L
af1
M0
60-L
c1L
af1
M0
80-L
c1L
af2
M0
60-L
c1L
af2
M0
80-L
c1L
af3
M0
60-L
c1L
af3
M1
00-L
as1
M0
80-L
as1
M0
60-L
as1
M1
00-L
as2
M0
80-L
as2
M0
60-L
as2
M1
00-L
as3
M0
80-L
as3
M0
60-L
as3
TC
O (
k€)
57
45
32
65
52
83
70
101
87
119
105
92
80
67
114
102
88
134
123
109
Table 4
Financial Analysis The financial analysis is the study of debt size and equity investment with which the developer will cover the wind farm costs. This leverage effect, as well as the resulting Internal Rate of Return (IRR), are the very important metrics the developer is willing to increase at most. The variation of leverage effect is studied with the AEP uncertainties previously calculated and with the following financial parameters:
Table 5
The table below summarizes the equity investment, debt size and IRR resulting from the different projects due diligence calculated with a financial model developed by Eurowatt Services, France:
Table 6
This table shows for instance that from an 80m met mast alone to an 80m met mast with a lidar planned at three locations and seasonally moved among these locations, the developer has saved 5.3M€ of equity investment through the reduction of the investment risks, when the wind farm project is presented to banks for obtaining the loan.
Best Measurement Strategies The best measurement strategies can be defined as the highest reduction of equity investment with the lowest operational costs (highest RoI). On the graph below, the 60m mast is taken as reference and strategies are shown as further operational costs (x-axis) and equity savings (y-axis) in comparison with the reference:
Figure 2
There are some financial conclusions that can here be drawn:
Lidar away from mast has higher RoI than lidar close to the mast
Every new location increases the equity savings, the RoI however decreases with
increasing number of locations
Lidar with seasonal moves has higher RoI than lidar at fixed locations
The equity savings can reach several millions of Euros with an extra expenditure of a few tens of thousands Euros invested in the energy yield assessment campaign.
Conclusion DEWI & LEOSPHERE study confirms that adding a highly accurate and mobile measurement system, like the certified WINDCUBE®v2 lidar, in an energy yield assessment leads to lower modeling uncertainties compared to having only one mast on site, or, compared to several masts on site to lower costs for the measurement campaign. Using a WINDCUBE®v2 in the assessment of the energy yield has therefore a high Return on Investment: it increases the wind farm value and considerably decreases the developer financial effort, saving millions of Euros in its equity investment.
References 1. M. Boquet & al., “Return on Investment of a Lidar Remote Sensing Device”, DEWI
Magazine #37, sept. 2010, pp.56 to 61 2. D. Faghani & al., Helimax - GL Garrad Hassan. “Remote Sensing: Practical Use for a
Wind Power Project”. AWEA Wind Resource Assessment Workshop 2009. 3. I. Campbell & al., “A Comparison of Remote Sensing Device Performance at Rotsea
site”, Document Reference: 01485-000090 Issue: 05 – Approved, RES Group 4. A. Albers & al., “Comparison of Lidars, German Test Station For Remote Wind Sensing
Devices”, Deutsche Windguard GmbH 5. M. Strack, W. Winkler, “Analysis of Uncertainties in Energy Yield Calculation of Wind
Farm Projects“, DEWI Magazine No. 22 (2003), pp. 52 to 62 6. C. Bezault et al., “Sensitivity of the CFD based LIDAR correction”, Meteodyn, EWEA
2011 7. D. Faghani & al., “LiDAR Validation in Complex Terrain”, GL Garrad Hassan, EWEA
2011