Environmental Engineering Reference
In-Depth Information
Table 15-2.
An example calculation of the array-average uncertainty for a layout of
65 turbines associated with five different masts. The arrows indicate the order of
calculation. First, the columns of correlated uncertainties are combined according to
Equation 15.8, and the columns of uncorrelated uncertainties according to Equation
15.7. Next, the total uncertainties in the bottom row are combined assuming the
individual components are uncorrelated.
Historical
Resource
(Correlated)
Future
Resource
(Correlated)
Other
Uncertainties
(Uncorrelated)
No.of
Turbines (N)
MAST
Total
A
15
3.0%
1.5%
5.5%
6.4%
B
7
3.0%
1.5%
6.2%
7.0%
C
20
3.0%
1.5%
5.8%
6.7%
D
9
3.0%
1.5%
6.0%
6.9%
E
14
3.0%
1.5%
5.6%
6.5%
Combined
65
3.0%
1.5%
2.7%
4.3%
Source
: AWS Truepower.
stub masts. Because of the acceleration of wind over the tower top, this introduced
a positive bias in both the observed wind speed and the wind shear, resulting in a
doubly serious overestimation of the projected hub-height wind speed. Similarly, it
used to be common practice to place towers at the very best—most exposed, highest
elevation—points within a project area. This, too, led at times to biased wind resource
estimates because of the tendency of wind flow models to underestimate the drop in
wind speed with decreasing elevation.
Given this history, the conservative analyst might choose to assume that all the
errors associated with one mast are correlated with those of other masts. This assump-
tion, however, takes no credit at all for deploying multiple masts at a project area
and could result in an overly pessimistic assessment of the financial risks in the wind
project. By judging each source of error on its merits and treating them in a consistent
way, it should be possible to arrive at an objective assessment of the wind resource
uncertainty.
15.7 QUESTIONS FOR DISCUSSION
1. What is uncertainty? Why is it important to consider the uncertainty band in a
given wind resource assessment analysis?
2. What are the primary factors that are typically considered when quantifying the
uncertainty of a wind resource assessment program? What are some specific
factors that determine the magnitude of each?
3. Why would averaging concurrent valid observations from two anemometers at
the same level reduce the overall measurement uncertainty?
Search WWH ::
Custom Search