University Physics Notes: Classical Mechanics – Introduction to Perturbation Theory
Real problems rarely have Hamiltonians with equations of motion having simple solutions depending upon elementary functions, so some method of approximating the solutions is needed. Typically we start with a Hamiltonian of the form
(1)
where the unperturbed Hamiltonian
is
soluble but the perturbed Hamiltonian (1) is not. The methods of
finding approximate solutions constitute perturbation theory. The
perturbation
may
be independent of the time, in which case the perturbed system is
conservative, or may depend on the time – the methods used to find
approximate solutions are different in each case.
The most famous perturbation problem is the motion of the planets.
The greatest influence on the motion of the planets is the Sun, with
each planet modifying the orbit of each other planet by only a small
amount. For example, the motion of Venus around the Sun is perturbed
most by Jupiter, whose mean force on Venus is less than
times
the force exerted by the Sun, and the Earth, which exerts an average
force less than
times
that of the Sun. As a first approximation, only the influence of the
Sun is considered. The Hamiltonian is one of central force type and
simple to solve. The effects of Jupiter and the other planets are
small perturbations to the motion.
Often the set of solutions to the unperturbed system H_0 form a
complete set,
so that any function, within limits can be expressed as a sum of
these solutions. This extends to the perturbed Hamiltonian, so that
an solution to the perturbed Hamiltonian can be expressed as a sum of
solutions to the unperturbed solutions. The problem then becomes
finding the coefficients $epsilon in the expansion. Finding the
solutions of equations may mean treating
as
the variable and finding the coefficients
of![]()
![]()
Perturbation theory is particularly useful when a system contains parameters and we need to know the effect of these on the solutions, and also when we want to analyse the qualitative behaviour of solutions. For example, is is relatively easy to solve the equation
![]()
for various values of
to
reasonable accuracy, but there are often circumstances when it is
useful to know that the root behaves as
where![]()