The Convolution Theorem for Laplace Transforms

 

Background. (Convolution Theorem) Let [Graphics:p20.txtgr1.gif] denote the Laplace transforms of [Graphics:p20.txtgr2.gif], respectively. Then the product [Graphics:p20.txtgr3.gif]is the Laplace transform of the convolution of [Graphics:p20.txtgr4.gif], and is denoted by [Graphics:p20.txtgr5.gif], and has the integral representation
[Graphics:p20.txtgr6.gif]

Load Mathematica's built in "LaplaceTransform" subroutine package.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr7.gif]
 

Computer Lab Work.

 

Example 1. Use convolution to find the inverse Laplace transform of [Graphics:p20.txtgr9.gif].
Solution: H(s) is the product of [Graphics:p20.txtgr10.gif],
which are the Laplace transoms of f(t) = sin(t) and g(t) = 2 cos(t), respectively.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr11.gif]
[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr12.gif]

We can check this with Mathematica's result using the InverseLaplaceTransform package.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr13.gif]
 

Example 2. Use the convolution theorem to solve the integral equation
[Graphics:p20.txtgr14.gif].
Solution: Take the Laplace transform of each of the terms and get:

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr15.gif]

Use Mathematica to find the inverse Laplace transform, and plot f[t] over the interval 0 <= t <= 10.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr16.gif]

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr17.gif]

 

More Background. The Laplace transform of the Dirac delta function [Graphics:p20.txtgr18.gif] is F(s) = 1. This can be illustrated with Mathematica.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr19.gif]
 

Example 3. Solve the initial value problem
[Graphics:p20.txtgr20.gif]with [Graphics:p20.txtgr21.gif].
Solution: Taking transforms results in the equation:

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr22.gif]

Use Mathematica to find the inverse Laplace transform, and plot f[t] over the interval 0 <= t <= 3.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr23.gif]

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr24.gif]


Example 4. Use convolution to solve the initial value problem
[Graphics:p20.txtgr25.gif]with [Graphics:p20.txtgr26.gif].
Plot the solution over the interval 0 <= t <= 1.57

Solution. First solve [Graphics:p20.txtgr27.gif] with [Graphics:p20.txtgr28.gif].

Second solve for [Graphics:p20.txtgr29.gif] with [Graphics:p20.txtgr30.gif].

Then the desired solution is y(t) = u(t) + v(t).

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr31.gif]
[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr32.gif]

Now form y[t] and plot the solution over 0 <= t <= 1.57

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr33.gif]

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr34.gif]

If you wish to plot the solution over a larger interval, then y[t] will need to be simplified.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr35.gif]

Furthermore, we need to take the logarithm of the absolute value, and this must be done by hand.

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr36.gif]

[Graphics:p20.txtgr8.gif][Graphics:p20.txtgr37.gif]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(c) John H. Mathews, 1998