Angle of Loll Calculation by Cubic Spline: Journal of Maritime Research
Angle of Loll Calculation by Cubic Spline: Journal of Maritime Research
Angle of Loll Calculation by Cubic Spline: Journal of Maritime Research
Article history: Several years ago the Basque Government supported the programming of the software ARKITSAS in order to provide
Received 28 February 2013; all existing vessels with a specific software to calculate stability, cargo and longitudinal strength data. The aim of this
in revised form 12 March 2013; article is to present the part of that research concerning the definition of the static stability curve by cubic spline in its
accepted 25 May 2013 initial end when the metacentric height is negative. Taking into account that the slope at initial end is known, the pre-
cision of the results for low heeling angles may be improved and, in this way, the accuracy in the calculation of loll
should be enhanced. This method of calculation is compared to other traditional methods used for wall-sided ships by
Keywords: the application to three different ships.
Static stability curve, Cubic splines,
Free and fixed end, Angle of loll.
© SEECMAR / All rights reserved
1. Introduction within the initial stability more precise and, taking into ac-
count that the angle of loll is usually small; its calculation
The static stability curve represents the values of GZ arms for would be suitable by means of non-free end cubic spline
the different heeling angles. However, the stability booklet method.
usually provides GZ arms for every ten or fifteen degrees of When the ship is ‘wall-sided’ the approximate formula in
heeling angles. This means that the rest of righting arms have (1) may be enough to calculate the angle of loll (Barrass and
to be obtained by drawing the static transversal stability curve Derret, 2012). However, this method of calculation would not
passing through the known data. In this way, the global cubic be suitable for those ships that have very fine bowlines and
splines seems to be one of the most suitable methods to define stern contours. Even in the case of box shaped ships the angle
the static stability curve since the same have to pass through of loll calculated by this formula is not as accurate as that ob-
some control points. The local splines pass also through the tained graphically from the static stability curve. For this rea-
control points, although its degree of smoothness is lower than son, the method presented in this paper takes advantage of the
the global splines. On the other hand, the B-splines would not graphic definition of the static stability curve by cubic spline
be suitable because the curve would not pass through the con- to obtain the angle of loll by the cross of this curve and the ab-
trol points. Consequently, the GZ arms obtained by global scissa axis.
cubic spline interpolation appear to be appropriate enough.
The ends of spline curve are usually free under this inter-
(1)
polation method. However, the slope of the static stability
curve for the heel of 0º (when the vessel is up righted) is known
specifically, it is equal to the metacentric height divided by the
value of one radian in degrees. Therefore, it is possible to fix 2. Static stability curve definition by cubic spline
the initial end of the static stability curve to make the GZ arms
The static stability curve defined by global cubic splines is
made up of different portions of curves connected in control
1 University of Basque Country. María Díaz de Haro, 68. 48920 Portugalete. Spain.
2 Professor, Email: imanol.basterrechea@ehu.es, Tel. +0034946014792, Fax. 34946017700. points called knots which coincide with the known data.
3 Professor, Email: iranzu.sotes@ehu.es, Tel. 34946014848, Fax. 34946017700.
Therefore the curve is a piece-wise function defined by mul-
4 Professor, Email: alberto.lopeza@ehu.es, Tel. 34946014832, Fax. 34946017700.
5 Professor, Email: inaki.alcedo@ehu.es, Tel. 34946014840, Fax. 34946017700.
tiple subfunction in the form of the equation in (2), n being
* Corresponding Author. the total data provided in the cross stability curves.
22 Journal of Maritime Research,Vol. X. No. 2 (2013)
(9)
(2)
where:
Obviously, the adjoining subfunctions coincide in value at
knots and, moreover, the slope and the curvature at knots is
the same for the adjoining subfunctions, which makes the
whole curve smooth. Therefore, the following conditions have
to be fulfilled (Borse, 1991):
1st condition: the value of a cubic subfunction at initial knot
is known.
(3)
(4)
(5)
(7) (11)
If these new coefficients B0 and C0 are applied to the first so obtained will be compared to those calculated by the ap-
four conditions, the values of proximate formula in (1). Obviously, the ships are unstable in
A1,1 and R1 will also vary. all cases to get a negative GM. On the other hand, the static
(14) stability curve is drawn by cubic splines either fixing the ends
or leaving them free.
FIRST CASE.- 16,600 dwt bulk-carrier (wall-sided).
Draft 10 meters
Taking into account the new values obtained from the for- TKM (vertical distance from keel to transversal
mulae (9) to (15), the equation in (9) is applied again to calcu- metacentre) 9.707 meters
late the values of Ci and the rest of coefficients. Thus, the first GM (metacentric height) 0.043 (-)
end of the static stability curve will be fixed. VCB (vertical center of buoyance) 5,305 meters
BM (metacentric radious) 4.402 meters
Source: Authors.
(17)
4. Practical application
Figure 2: Small angles’ detail of the GZ curve for 16,600 dwt bulk- Table 6: GZ arms for sailing yacht, draft 2.71 meters and GM 0.05(-) meters.
carrier (wall sided).
φ 0° 10° 20° 30° 40° 50° 60°
GZ 0 0.0096 0.032 0.02 -0.015 -0.1246 -0.244
Source: Authors.
Source: Authors
In case that the hydrostatic data and the GZ arms are those
showed in table 3 and 4, respectively, the coefficients of first
spline subfunction are expressed below.
B0 = -3.4904∙10-4
C0 = -2.4409∙10-4 Source: Authors
D0 = 5.0899∙10-5
The angle of loll is thus calculated from the equation in (16). THIRD CASE.- Sailing yacht (round-shaped)
φloll = 5.95°
On the other hand, the angle of loll obtained from the for- The coefficients obtained from the hydrostatic data and
mula in (1). the GZ arms from the table 5 and 6, respectively, are shown
φloll = 3.33° below:
B0 = -8.7260∙10-4
Table 5: Sailing yacht hydrostatic data.
C0 = 1.9003∙10-4
Draft 2.71 meters D0 = -6.7682∙10-7
TKM (vertical distance from keel to transversal
The angle of loll obtained from the equation in (16).
metacentre) 3.35 meters
GM (metacentric height) 0.05 m. (-)
φloll = 4.67°
VCB (vertical center of buoyance) 1.94 meters On the other hand, the angle of loll obtained from the for-
BM (metacentric radious) 1.41 meters mula in (1).
Source: Authors. φloll = 14.9°
I. Basterretxea, I. Sotés, A. López and I. Alcedo 25
Figure 5: GZ curve for sailing yacht (round-shaped). Figure 6: Small angles’ detail of the GZ curve for sailing yacht (round-shaped).
5. Conclusions that the approximate formula is not valid for non wall-sided
ships.
The results obtained in the practical application show that the The global cubic splines let us obtain graphically and ana-
approximate formula is not accurate enough for round-shaped lytically the angle of loll in an accurate way. However, it is es-
ships, although it may be useful for wall-sided ships. The re- sential to fix the end where the slope of the static stability
sults of the applications of both methods in the 16,600 dwt curve is known. Otherwise, there may be an error of more than
bulk-carrier In the case of the 150,000 dwt tanker, there is a three degrees. Therefore, it is advisable to fix the end when a
difference of almost three degrees between both methods of math program such as Matlab or Mathematica is used to de-
calculation, which seems to be excessive for a box-shaped ship fine the static stability curve.
with small block coefficient; nevertheless, it must be taken into
account that the ship’s conditions to reach a negative meta-
centric height have been forced too much. This is due to the References
fact that the shape and hydrostatic particulars of that type of
ship provide her with excessive stability. Therefore the height Barrass, C.B. and Derrett, D.R. (2012) Ship stability for Masters and Mates. Oxford:
Butterworth-Heinemann.
of the centre of gravity estimated for the tanker in the practical Borse, G.J. (1991) FORTRAN 77 and numerical methods for engineers. Boston: PWS
case is hypothetical and out of the possible stability criteria in Publishers.
the construction of this type of ships (Riola and Pérez, 2009). Rawson, K.J. and Tupper, E.C. (2001) Basic ship theory. Oxford: Butterworth-Heine-
mann.
In the case of the sailing yacht, the difference of the results be- Riola, J.M. and Pérez, R. (2009) Warship damage stability criteria case study. Journal
tween both methods is bigger than ten degrees, which means of Maritime Research 6 (3), 75-100.