The following is a list of the major topics which will be covered on this exam. It is not presented in the same order as the textbook.
Here are more detailed descriptions of some of these ideas and concepts.
(1) Generating and intepreting a mathematical model for a
specific physical situation is one of the more difficult topics we
considered.Most of the types of problems we considered led to simple
separable or linear equations. You should know how to solve an
acceleration-velocity problem with or without air
resistance and understand the different possibilities for modelling air
resistance, i.e., linearly or quadratically proportional to the velocity.
(2) The two numerical methods we studied were Euler's
method and the Runge-Kutta method. You should be able to
use these methods to derive approximations to solutions using a
calculator. I will only ask you to iterate the method
one or two times. Due to its complexity, I will provide you with the
formulas for the Runge-Kutta method.
(3) Recognize general homogeneous and
nonhomogeneous linear equations. The general solution to an
nth-order homogeneous linear equation is a linear combination
of n linearly independent functions. Know what it means for a
collection of functions to be linearly dependent or linearly
independent and be able to use the Wronskian to test this.
Know how to solve constant coefficient equations using the
characteristic equation. Remember that there are various cases,
according to the multiplicity of the roots of the characteristic
equation and whether those roots are real or complex.
The general solution y(x) to a nonhomogeneous linear equation
is the sum of the complementary solution y_c(x), which
is the general solution to the associated homogeneous equation,
and one particular solution y_p(x) to the given equation. We
discussed two methods for finding the particular solution
y_p(x): the method of undetermined coefficients and the
method of variation of parameters. The former is easier to
apply but only works in certain special cases; the latter works all
the time but reduces the problem to the computation of a (possibly
intractable) integral. Regarding the method of undetermined
coefficients: know how to choose the right guess for y_p(x).
It should be a linear combination of all of the terms which appear in
the forcing term of the original differential equation or any of its
derivatives. The situation is somewhat more complicated if any of
these terms satisfy the associated homogeneous equation; then it is
necessary to multiply all of the terms in your guess by a suitable
power of x.
Here are a few sample problems for you to try:
Solutions.