Trans RINA, Vol 152, Part B1, Intl J Small Craft Tech, 2010 Jan-Jun
3.1 REAL-TIME VPP A Real-Time VPP for wind tunnel
testing has been
developed for use in the Twisted Flow Wind Tunnel as described in detail by Hansen et al. [5]. It was designed as an extension to FS-Equilibrium (a semi-empirical VPP developed by FutureShip GmbH). The wind tunnel forces are measured by the Real-Time VPP LabVIEW application and passed to FS-Equilibrium in coefficient form where the force equilibrium is calculated for up to six-degrees of freedom based on the aerodynamic wind tunnel data and standard semi-empirical hydrodynamic force models. The Real-Time VPP makes the trimming of the sails more realistic because the resulting changes in boat speed, heel angle, rudder angle, etc. can be seen in the wind tunnel during the actual testing. While the sails are trimmed the measured forces and the predicted performance is
constantly updated at a rate of
approximately 5Hz. Once the desired trim is obtained the forces and moments are measured for a longer period (in this study 60 seconds) and the mean values are recorded together with the resulting
performance. Another
advantage of the Real-Time VPP is that it calculates the resulting heel angle of the yacht so that it is now possible to dynamically rotate the model in the wind tunnel to the correct heel angle, instead of having to conduct the tests at a number of predefined fixed heel angles.
3.2 TESTING PROCEDURE
In this study the model was not dynamically heeled to ensure that
the experiments produces only force
differences due to the trimming of the sails. If the tests are conducted at different heel angles the effect of heel on the sail forces would be include in the results, which could be much greater than the differences in the trim parameter models. It is also thought that the process of de-powering is not influenced significantly by changes in heel angle. The tests were therefore conducted at 0° heel and in the VPP calculations effective angle theory was used to account for the heeling of the yacht. Tests were conducted with the mainsail and jib for six apparent wind angles of 20°, 25°, 30°, 40°, 60° and 90°. For each apparent wind angle the sails were trimmed using the Real-Time VPP by maximising the boat speed for true wind speeds between 2m/s and 14m/s. The trim for the true wind speeds of 2m/s and 4m/s represents the fully powered up base trim. For higher wind speeds a higher boat speed is achieved by de-powering the sails at certain apparent wind angles. With the Real-Time VPP these depowered trims can be obtained efficiently. The force equilibrium is solved for all six-degrees of freedom.
3.3 MODELLING OF DATA WITH TRIM PARAMETER POWER
It is not sufficient to use the measurements obtained with the Real-Time VPP directly for a comparison with the trim parameter models because experiments inevitably produce point measurements and not a continuous data
set. In order to assess the differences in de-powering accurately it is also important that powered-up input
data is used. The
identical fully changes in
performance due to different trim parameter models are small compared to differences that can arise from curve fitting of the fully powered-up data. The wind tunnel data is therefore post-processed. The fully powered-up data is used to produce the fully powered-up input curves, which are used for both the experimental and the generic trim parameter models. The depowered wind tunnel data are used as input for the empirical ‘power’ model, which describes de-powering in detail while smoothing out some of the inaccuracies in the individual measurements as described by Hansen et al. [6]. The ‘power’ model employs the fully powered-up input curves and response surfaces that describe de-powering as functions of βeff and the trim parameter power, which is a measure of the reduction in heeling moment. The lift coefficient (CL), for example, is calculated from the optimum fully powered-up lift coefficient (CLopt) and the ‘power’ ratio of lift (RL) so that () (
CC ß L = Lopt eff L effβ , R power) . (8)
Figure 3 shows the CL measured in the wind tunnel with the Real-Time VPP and the CL obtained when modelling the measurements with equation (8). The modelled CL is only calculated for each wind tunnel measurement and the points are connected by straight lines since the power
values at other βeff are not known. CL is modelled in good detail but some of the discontinuities due to measurement inaccuracies are smoothed out. The average absolute error between the CL measurements and the response surface is 1.2% and the maximum absolute error is 4.5%.
2.8 2.4
VT=14m/s RT-VPP VT=14m/s modelled VT=12m/s RT-VPP VT=12m/s modelled VT=10m/s RT-VPP VT=10m/s modelled
2
VT=8m/s RT-VPP VT=8m/s modelled VT=6m/s RT-VPP VT=6m/s modelled VT=4m/s RT-VPP VT=4m/s modelled
1.6
1.2
0.8
0.4
0 0 153045 βeff [°] 607590
Figure 3: Lift coefficient (CL) for all tested true wind speeds (VT) from Real-Time VPP and modelled by power
B-12 ©2010: The Royal Institution of Naval Architects
CL [-]
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60