origin along the line Ox (Fig. 2.12b). The ellipses
and hyperbolae are called confocal because they
share common foci located at x ϭϮf.
Given an arbitrary point P(␰, ␩) in elliptical
coordinates, the Cartesian coordinates are found
using the following transformation equations
(Timoshenko and Goodier, 1970):
(2.32)
The functions cosh and sinh are hyperbolic functions. Just as in the polar coordinate system where
any circle centered at the origin would be represented by a constant value of r, each of the family
of confocal ellipses is represented by a constant
value of ␰. However, unlike the coordinate r, which
is a length measured in meters, the coordinate ␰
is just a number. Also, as each radial line in the
polar system is associated with a particular value
of the angle ␪, each of the family of confocal
hyperbolae is associated with a particular value of
the angle ␩.
x ϭ f cosh ␰ cos ␩, y ϭ f sinh ␰ sin ␩
The position vector, p, used to locate the point
P(␰, ␩) in the two-dimensional elliptical coordinate system (Fig. 2.12b) is written as a linear combination of the scalar components (2.32) and the
two mutually orthogonal unit base vectors (e x , e y )
of the Cartesian system:
(2.33)
The scalar components carry the units of length
(meters) because f has units of length. Thus the
focal length determines the length scale of this
coordinate system.
For a given hyperbola (␩ ϭ constant), as ␰
ranges from numbers much less than one to
numbers greater than one the ellipses range from
very eccentric to nearly circular. In the limit, as ␰
→ 0 the ellipse collapses down onto the x-axis,
extending from one focus to the other. Later we
use such highly eccentric ellipses as models for
fractures and faults in rock. For a given ellipse (␰ ϭ
constant), as the angle ␩ ranges from 0Њ to 360Њ,
points vary in position around the ellipse in a
counterclockwise direction starting at the intersection with the positive x-axis where ␩ ϭ 0Њ. The
angle ␩ ϭ 90Њ at the intersection of the ellipse and
the positive y-axis; ␩ ϭ 180Њ at the intersection
with the negative x-axis, and so forth.
The semi-major axis, a, and semi-minor axis, b,
of a particular ellipse (␰ ϭ constant) are:
(2.34)
For segment 16 of the Ship Rock dike (Fig. 2.12a)
we have a ϭ 68 m and b ϭ 1.7 m, so the particular
ellipse, ␰ O , defining the contact is:
(2.35)
The specification of boundary conditions for problems related to dike segments is particularly
simple using elliptical coordinates. For example,
if the magma pressure, p m , exerts traction, t, only
perpendicular to the contact, we write:
(2.36)
Here t ␰ is the traction component acting on the
elliptical surface, ␰ O , in the ␰-coordinate direction
BC: on ␰ ϭ ␰ O Ά
t ␰ ϭ p m
t ␩ ϭ 0
␰ O ϭ tanh Ϫ1
΂
b
a ΃ ϭ tanh Ϫ1 (0.025) ϭ 0.025
a ϭ f cosh ␰, b ϭ f sinh ␰
ϭ ( f cosh ␰ cos ␩)e x ϩ ( f sinh ␰ sin ␩)e y
p ϭ p x e x ϩ p y e y
2.2 LOCAL COORDINATES AND POSITION VECTORS
47
(b)
x
y
f
P(j, h)
j = constant
h = constant
a
b
(a)
j = j O
p
O
Fig 2.12 (a) Aerial photograph of one segment of the
northeastern dike at Ship Rock (Delaney and Pollard, 1981).
(b) Elliptical coordinate system used in models of dike
segments.
Précédent

- 61/516

Suivant