Scribd
Upload
254 views16 pages

Ischemia and Loss of Vascular Autoregulation in Ocular and Cerebral Diseases A New Perspective Verified Download

The document discusses the phenomenon of bifurcation in various physical systems, highlighting its mathematical underpinnings and applications in fluid dynamics, particularly in the context of convection. It emphasizes the importance of stability and the emergence of new solutions as parameters change, illustrated through examples like the Benard problem. Additionally, it touches on the implications of bifurcation in geophysical and astrophysical contexts, such as the Earth's magnetic field generation.

Uploaded by

goongt.ynhphuy.nhtach
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
254 views16 pages

Ischemia and Loss of Vascular Autoregulation in Ocular and Cerebral Diseases A New Perspective Verified Download

The document discusses the phenomenon of bifurcation in various physical systems, highlighting its mathematical underpinnings and applications in fluid dynamics, particularly in the context of convection. It emphasizes the importance of stability and the emergence of new solutions as parameters change, illustrated through examples like the Benard problem. Additionally, it touches on the implications of bifurcation in geophysical and astrophysical contexts, such as the Earth's magnetic field generation.

Uploaded by

goongt.ynhphuy.nhtach
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

Ischemia and Loss of Vascular Autoregulation in Ocular and

Cerebral Diseases A New Perspective

Visit the link below to download the full version of this book:

https://medipdf.com/product/ischemia-and-loss-of-vascular-autoregulation-in-ocul
ar-and-cerebral-diseases-a-new-perspective/

Click Download Now


a differential equation by algebraic methods.

I would like to thank the Mathematics Department of the

University of Chicago for their kind hospitality, Peter Olver for his

useful remarks during the lectures and his contribution to these notes,

and F. Flowers for an excellent job of typing.

My sabbatical was supported by the University of Minnesota,

the National Science Foundation (MCS 73-08535) and the U.S. Army

Research Office (DA AG 29-77-G-0122), whose support is much

appreciated.

David H. Sattinger June 1978


TABLE OF CONTENTS

I PHYSICAL EXAMPLES OF BIFURCATION ....................

II MATHEMATICAL PRELIMINARIES .......................... 18

III STABILITY AND BIFURCATION ........................... 37

IV BIFURCATION AT MULTIPLE EIGENVALUES ................. 70

V ELEMENTS OF GROUP REPRESENTATION THEORy............. 96

VI APPLICATIONS ........................................ 151

VII

APPENDIX: HOW TO FIND DIFFERENTIAL EQUATION (by Peter


Olver)

THE SYMMETRY GROUP OF A


200

Subject Index

240

PHYSICAL EXAMPLES OF BIFURCATION

If a layer of fluid is heated from below, convective instabilities

set-in when the temperature drop exceeds a certain critical value,

and the convective motions which evolve often display a striking

cellular structure, as pictured below.

Figure 1. 1 + Convection cells in the Benard problem [12]

+ E. Koschm ieder: Benard Convection. Advances in Chemical Reprinted


from: Physics 26 , 177 (1974) by permission of John Wiley and Sons, Inc.
2

(See Chandrasekhar [4] or Koschmieder [11], [12])

The onset of such

convective motions provides a striking physical example of the phenome-

non of bifurcation - - that is, the appearance of multiple solutions to a

nonlinear system of equations.

To state the situation mathematically, let us suppose that the

equilibrium states of a physical (electrical, mechanical, chemical, bio-

logical, what have you) system are determined as solutions of a functional

equation
G(x.,u) = 0 (1.1)

where X. is a parameter, u is an element of a vector space

(for example

a Banach space) and G is a mapping from

to another vector space .J. .

We shall give specific examples of such equations later on. Let us suppose

that u = 0 is always a possible solution of (1.1), that is, G(x.,O) EO.

A bifurcation point

(x. ,0) is a critical value of X. at which several solution c

branches of (1.1) confluesce.

For example, there may be a one-parameter


family of nontrivial s(!)'lutions

x. = X.(E), U = U(E) , G(X.(E), U(E)) ;: 0 ,

x.(0) = X.c

u( 0) = 0 .

The situation is indicated schematically in Figure 1.2 .

x. (t ), u( E )

>-.

Figure 1.2
Bifurcation point (x.c,O); one non-trivial branch. The vertical axis
represents the vector space

Closely tied to the phenomenon of bifurcation is the property of

stability . Roughly speaking, a solution is stable if small perturbations

from it remain small (We are talking now about the dynamical situation).

A solution is asymptotically stable if small disturbances decay to zero in

time. When external parameters are varied (such as the temperature

drop across the layer of fluid in the Benard problem), one solution may

become unstable as the parameter crosses the critical value, and it is

at such a transition point that new solutions bifurcate from the known

solutions.
The new 'Solutions which appear at the transition to instability need

not be stationary solutions. They may be time-dependent solutions -

for example time-periodic solutions. In such a case, the loss of stability

manifests itself as the appearance of oscillatory motions; generally, the

amplitude of these motions increases with the increasing parameter value.

The appearance of time-periodic motions is especially important in the


theory of simple mechanical systems (two degrees of freedom) such as

clocks and vacuum tube circuits.

E. Hopf in 1942
[7] gave a general theorem for the bifurcation of periodic motions in
systems with n degrees of freedom. He also discussed the possible
extension of his results to systems with an infinite number of degrees of
freedom (i. e., systems governed by partial differential equations) and
hydrodynamics in particular. Hopf's results have now in fact been
generalized to the Navier-Stokes equations (the partial differential equations
governing the dynamics of a viscous incompressible fluid) and more
generally to parabolic systems of partial differential equations. The
generalization to partial differential equations has been discussed in
Marsden and McCracken's book [14] and in my lecture notes [18]. I will
give a partial derivation of the results in Chapter III of these notes. The
Benard problem (convection in a plane layer) mentioned above is a model
problem which has been much studied in the laboratory because it has given
such insight into the processes of bifurcation in fluid

dynamics.

It is a sensitive problem and many fundamental questions

remain unresolved to this day.

Of more importance to the physical

sciences, however, is the onset of convection in a spherical geometry. These


questions have an important bearing on problems in geophysics and
astrophysics where the onset of convection in a star or in the earth's
5

core are of interest. We shall discuss these and other bifurcation problems

of physical interest below; but the list is only indicative of the variety of

bifurcation phenomena which occur in nature.

1. The equations employed to model convection are the so-called

Boussinesq equations [4]

uk + 0k3 e

aX k

1 aUk -u-+ iP r j ax. J


a

at

(1. 2a)

e+

u3=

ae u- j ax. J

+ ae at

(1.2b)

au. --.L = 0 ax. J

(1. 2c)

3 where

= VIK is the Prandtl number, and "Ie. = av


h (T o - T 1 ) is the

Rayleigh number. Here g is the acceleration due to gravity, h is a

characteristic length (for example, the depth of the layer of fluid), a is

the coefficient of thermal expansion of the fluid; TO and T 1 are the


tempera-

tures on the lower and upper boundaries; v and K are the coefficients of

viscosity and thermal conductivity; e is the perturbed temperature profile;

and u 1 ' u 2 ' u 3 ' p are the components of velocity and pressure. 0k3 is the

Kronecker delta function.

These same equations are also valid in a spherical geometry. In

geophysical applications where the body is rotating, an additional term due


to Coriolis forces must be added to the right-s.ide of the first equation.

In vector form this term is 2Q X u where Q is the rotation vector

(£klm Q 1 u m in component form) .

One of the outstanding problems in geophysics is the Dynamo

problem: How is the earth's magnetic field maintained? The consensus

now is that the magnetic field is sustained by dectric currents flowing in

the electrically conducting earth's core.

It is known that purely axi-

symmetric fluid motions cannot sustain a magnetic field; but if convective


motions accounted for asymmetric fluid motions, these might sustain such

an electromagnetic field. The equations of dynamo theory are the con-

vection equations plus an equation for the magnetic field:

au + (u. v) u + 2Q X u at

1 - VP- I3ge + v

u + -(V X E) X B ....

(1.3a)

di v

=0

(1.3b)

ae ( _ ) - + u'V e = at
-u.VT +K

eo

(1.3c)

aE at

= curl(uXB + 'I1

(1. 3d)

where TO is the base temperature profile in the absence of convection. The


term 2Q X \r is the Coriolis term due to rotation.

When no magnetic field is present the quadratic term (V X B) X B

in (l.3a) vanishes and we have a pure convection problem. When the con-

vective velocity \r rises to a magnitude and configuration which can sustain

You might also like