Perturbation theory has applications in a lot of different areas than quantum mechanics.
One well-known example is the free-surface problem in fluid mechanics (potential flow):
We seek a solution of Laplace's equation (i.e, the continuity equation rewritten in terms of the velocity potential) which satisfy the non-linear free surface conditions (fluid pressure equals air pressure, and kinematic condition) and horizontal bottom.
In particular, we seek a solution where the surface profile can be described as a dominantly monochromatic harmonic wave.
The non-linear corrections to the dominantly linear solution can be found by perturbation theory.
In short, perturbation theory is an indispensable tool for the analysis of systems whose behaviour is desribed by diff.eq's where we cannot find analytical solutions (i.e, most diff.eq's)
As chroot observes, (regular) perturbation theory is at its most effective where effects/behaviours lie in well-defined layers of importance (in an asymptotic limit). Then, we can peel off layer after layer by perturbation theory to gain a better approximation.
A host of specialized techniques has been developed, for example, WKBJ-method, Poincare-Lindstedt method, multiple scale analysis, and so on.