Professional Documents
Culture Documents
Submarine Dynamic Modeling: L D M Lδf
Submarine Dynamic Modeling: L D M Lδf
Abstract
This paper discusses the development of a
dynamic model for a torpedo shaped sub-
marine. Expressions for hydrostatic, added
mass, hydrodynamic, control surface and pro-
peller forces and moments are derived from
first principles. Experimental data obtained
from flume tests of the submarine are inserted
into the model in order to provide computer
simulations of the open loop behavior of the
system.
Figure 1: Torpedo shaped AUV: fully assembled
Introduction
Autonomous underwater vehicles (AUV) are finding
use in a variety of oceanographic survey applications
(Foresti [2001] Dhanak [2001]). These vehicles present
a comparatively low cost technology for underwater
exploration with a freedom of motion superior to
tethered-towed submersibles.
An experimental AUV (Figures 1 and 2) has been
designed by Reid [2001] at Queensland University of
Technology and the computer control system was in-
stalled by staff at the CSIRO, CMIT Automation
Group. This AUV is torpedo shaped, approximately
1.5m long × 150mm diameter. Four, independently Figure 2: Torpedo shaped AUV: casing removed
actuated, orthogonal, stern planes are used to control
its attitude. Power to the DC motor, which drives the
585mm wide . These experimental procedures, de-
propellor, comes from a battery supply via a current
scribed in detail in the Appendix, have provided data
amplifier, used to control the shaft speed.
regarding relationships between angle of attack (α) and
The mathematical dynamic model, described in this
the coefficients of lift and drag (CL , CD ), moment co-
paper, provides a useful tool for the understanding and
efficient (CM ) and control surface effectiveness (CLδf ),
tuning of the control system which automatically con-
plotted in this paper.
trols the attitude and depth of the submarine. This
work tracks similar developments described by Nahon
[1993, 1998] Equations of motion
Numerical values for the hydrodynamic coefficients Figure 3 shows the body frame of reference in which the
which are contained within the mathematical model equations of motion are written. Origin Cb is located at
have been evaluated, where possible, from experimen- the centre of buoyancy and the centre of gravity lies at
tal data derived from the full size AUV or a half size the point rG =[xG ,yG ,zG ]T . The components of rG are
model, inserted into an open flowing channel (flume), small since the submarine is deliberately designed to
Cb Xsub Evaluation of Forces and moments
Surge: u Five sets of forces/moments act on the hull:
Roll: p
(i)Hydrostatic forces
Ysub The orientation of the body frame relative to the world
Sway: v frame is described by Euler angles rotated in the order:
Pitch: q roll φ, pitch θ, yaw ψ.
Heave: w The static forces, weight (W) and buoyancy (B) act
Yaw: r through the centre of gravity and centre of buoyancy
respectively. When resolved onto the submarine body
Zsub frame, these become:
V.
1.5
D = 12 ρCD Af V 2 (6)
1
CD is related to the angle of attack (α) through a
parabolic relationship, plotted in Figure 4. 0.5
CD = aα2 + bα + c. (7) 0
' − 21 ρAf CD (u2 + w2 )α Lift force and pitching moment, when viewed in the
XZ-plane are derived from equation 12:
Similarly, drag force, when viewed in the XY-plane, Zl = − 12 ρAf CLα (u2 + w2 )αcosα
may be resolved into the Xsub - and Ysub - directions. (13)
Ml = 12 ρAf CMα (u2 + w2 )α
0.04 4
0.03 2
Lift (N)
0.02 0
0.01 −2
0 −4
−0.01 −6
−10 −5 0 5 10 15 20 25 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
deg. δ (rad)
r
Figure 6: Hull Moment: Cm versus α, V=0.5m/sec. Figure 7: Fin Lift: Lift vs δr , V=0.81m/sec.
Slope: Cmα =0.0974 rad.−1 NB: |slope| = 2 |Yuuδr | V 2 =-25 N.rad.−1
Lf in = 12 ρCLδf Sf in δe ve2 2
(16)
Mf in = xf in Lf in
1
where CLδf is the rate of change of lift coefficient wrt
fin effective angle of attack and Sf in is the fin planform 0
Zs = − 12 ρCLδf Sf in u2 δs− uw + xf in uq
U
Ms = − 21 ρCLδf Sf in xf in u2 δs − uw + xf in uq [1980] uses the non dimensional advance ratio J = nd
15
0 δ =−5
s
10
−0.5
5
−1 δ = −1
s
z
−1.5
δ =1
s
−2 −5
−1 −0.8 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 0.8 1
pulse width modulated (%)
−10 δs = 5
Summation of forces and moments Figure 10: Simulated trajectory at various stern plane
Combining like cross-terms from equations 4, 5 10,15, angles V=1.54 m/sec.
18 and 19 , we get the following hydrodynamic forces:
Yuv = Yuvl + Yuvf + Yuvd Trajectory in the Earth−fixed frame
Yur = Yura + Yurf 8
Zuw = Zuwl + Zuwf + Zuwd
Zuq = Zuqa + Zuqf 6
(22)
Muw = Muwa + Muwf + Muwl
δ =−1
Muq = Muqa + Muqf 4 δ =−3
r
r
P
Xext = XHS + Xu|u| u |u| + Xu̇ u̇ + Xuv uv + Xuw uw+ −2
Xv|v| v |v| + Xvr vr + Xw|w| w |w| + Xwq wq+
Xqq qq + Xrr rr + Xprop −4 δ =+3
r
P 2
δ =+1
Yext = YHS + Yuuδr u δrtop + δrbottom + Yur ur+ −6
r
References
0 Foresti G.L. [2001] Visual Inspection of Sea Bottom
Structures by an Autonomous Underwater Vehicle,
IEEE Transactions on Systems, Man and Cybernetics
−5
0 1 2 3 4 5 6 7 8 9 10 - Part B: cybernetics, Vol. 31, No. 5, October 2001.
time (s)
Dhanak M.R, An P. E., Holappa K. [2001] An AUV
Figure 12: Simulated roll vs time at various rudder Survey in the Littoral Zone: Small Scale Subsurface
angles. Variability Accompanying Synoptic Observations of
Surface Currents, IEEE Journal of Oceanic Engineer-
ing, Vol. 26, No. 4, October 2001.
degrees.
Reid S [2001], The design of an Autonomous Under-
x(t) water Vehicle, Undergraduate Thesis, Faculty of Built
15
Environment and Engineering , Queensland University
10 of Technology, 2001.
x (m)
0
proceedings, 1993.
Blevins R. D. [1984], Applied Fluid Dynamics Hand-
−1
0 5 10 15 20 25
−z(t)
30 35 40 45 50 book, Van Nostrand Reinhold Co., 1984.
time (s)
4 Newman J.N. [1980], Marine Hydrodynamics (third
edition), MIT Press, 1980.
−z (m)
2
Chambers S. [2001] , Dynamics of an Autonomous
Underwater Vehicle, Undergraduate Thesis, Faculty of
0
0 5 10 15 20 25 30 35 40 45 50 Built Environment and Engineering , Queensland Uni-
time (s)
versity of Technology, 2001.
Figure 13: Simulated trajectory as the unpowered sub- Morrison A., Yoerger D. [1993] , Determination of
marine surfaces. the Hydrodynamic Parameters of an Underwater Ve-
hicle During Small Scale, Nonuniform, 1-Dimensional
Translation, IEEE proceedings, 1993
Conclusions
A rigid body dynamic model discussed in this paper
has produced a set of results for the open loop be-
havior of the submarine under the action of rudder,
stern plane and motor inputs. The next phase of the
project is to implement closed loop automatic control
on depth, heading and roll angle of the submarine.This
will be discussed in a forthcoming paper.
Acknowledgements
The authors wish to acknowledge the support of
Queensland University of Technology, Schools of Me-
chanical and Civil Engineering who manufactured the
submarine and provided the experimental test facili-
ties. Design of the submarine and experimental fa-
Appendix: Experimental Data
Experimental data from which the submarine’s hydro-
dynamic coefficients were derived came from tests car-
ried out in an open flowing channel (flume). Figure 14
shows a half size scale model, suspended in the channel
by a vertical shaft attached to a three axis dynamome-
ter, capable of measuring forces in the x- (drag) and
y- (lift) directions and a pitching moment about the
z-axis.
Figure 16:
Fy Mz
CLδf = 1 2 , CMδf = 1 2 (25)
2 ρAf V 2 ρAf V L
Y
X
Channel
Figure 15: