Environmental Engineering Reference
In-Depth Information
For this rotating test, strain gages were placed on the blades and central column, using a
double-active gage in the bridge so that bending about a single axis would be measured.
Gages were placed near the blade roots on leading and trailing edges (to measure chordwise
bending), on the inside and outside surfaces at maximum blade thickness (to measure flat-
wise bending), and on the central column (to measure in-plane and out-of-plane bending).
On-board amplifiers boosted the strain signal to 2,000 psi/V for all gages, in order to give an
adequate level for passing through the sliprings and the long wires that are typically required
for modal tests on operating wind turbines. Because of expected difficulties in understanding
the modes of the rotating central column, two piezoresistive accelerometers were attached to
the upper column to record in- and out-of-plane motions.
Excitation
The scope of methods used to excite vibrations in wind turbines for modal analysis is
broad. These methods fall into the four general categories of human, impact, step-relaxation,
and natural-wind. As mentioned previously, the FFT-testing procedure allows flexibility in
the excitation method, which is of significant benefit.
Human
Human excitation is the simplest approach to implement, but has the most limitations.
The large displacement, low frequency modes of large wind turbines allow a test engineer to
manually input a quasi-sinusoidal force into the structure to excite a single mode of the sys-
tem. The frequency and damping parameters of the mode can be determined as the systems
rings down after the excitation is removed. This approach was used to test over 20 commer-
cial 17-m VAWTs in as little as 1ΒΌ hours each. This study also provided valuable insight
into the structural-dynamic variability of duplicate units in the field [Lauffer 1986].
Impact
Impact excitation methods range from using an instrumented 3-lb hammer to strike the
2-m VAWT at various locations on its blades and central tower to swinging a 1,000-lb ram
from a crane to strike the tip of the NASA/DOE 2.5-MW Mod-2 HAWT [Boeing 1982]. In
both cases, the impact force must be of sufficient amplitude and duration to excite the modes
and frequencies of interest. This is accomplished by prior calibration tests.
Impact testing was also one of the techniques used to test the Sandia 34-m VAWT Test
Bed in various stages of assembly. This allowed model correlation of substructure models
before final model assembly. The substructures tested included two blade sections and a
blade assembly. Step-relaxation and natural wind excitation were also used for other sub-
structures, as appropriate [Carne et al. 1989].
Impact testing was also used to study the use of a Laser Doppler Velocimeter (LDV)
for non-contact measurements of a HAWT in the field. This study used impact excitation to
compare the LDV to traditional accelerometers as well as to study damage detection tech-
niques. A comparison to natural wind excitation was also performed. The test object in this
study, a 15 m diameter HAWT in its parked configuration [Rumsey et al. 1997].
Search WWH ::




Custom Search