Neuro-Fuzzy Modeling and Fuzzy Control of a Fast Ferry - Fuzzy Modeling and Control: Theory and Applications - page 187

Fig. 10.4 Membership functions of the outputs of the fuzzy controllers

If SSN is 6 and Ucos μ is HF, MF, LF then Moment_correction is
Small otherwise is None
If SSN is 7 and Ucos μ is HF, MF, LF, M, H, VH then
Moment_correction is Small otherwise is None

The fuzzy phase controller has been defined in the same way. The controller
changes the oscillation frequency by changing the angle of the actuators. When the
angle is increased, the oscillation frequency is reduced; decreasing the angle increases
the oscillation frequency. Therefore the inputs are:
•

360] ( ◦ ), defined by 5 labels (Very Advanced, Advanced,
Correct, Lagged, Very Lagged).

δ o =[ − ,

Phase error, 360

The input angle ( ◦ ): N (none), MP (very small), P (small), M (medium), A (high)
and MA (very high)

•

.The
labels for the output angle mean: MP (very small), P (small), M (medium), A (high)
and MA (very high) (Fig. 10.4 , right). The rules of the phase controller are defined
in the same way than the amplitude controller, that is,

The output of this controller is the working angle that must be applied, ϕ

If input_angle is N and Phase_error is any then Output_angle is N
If input_angle is MP and Phase_error is Very_Advanced then
Output_angle is MP
If input_angle is MP and Phase_error is Advanced then Output_angle
is P
...

For instance, Fig. 10.5 shows the actuators operating at the encounter frequency
for Sea State code 6 (wave height 3.85m), ship speed 20 knots. Dashed line

represents the flap motion and the solid line is the T-foil. The total pitch moment has
been notably reduced using these fins.
These fuzzy controllers has been simulation tested for different sea states and ship
speeds (Santos et al. 2005a ). The results in regular waves are encouraging, there is a
considerable reduction of the vertical acceleration.

≈