Laplace Transform Within a Domain

Click For Summary

Homework Help Overview

The problem involves finding the Laplace transform of a piecewise function defined as f(t) = t for 0 ≤ t ≤ T and 0 otherwise. The discussion centers around the application of Laplace transform properties and the handling of unit step functions.

Discussion Character

  • Exploratory, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the representation of the function using unit step functions and the implications for taking the Laplace transform. There are attempts to manipulate the integral form of the Laplace transform and questions about evaluating it over a restricted domain.

Discussion Status

Some participants have provided insights into the transformation process and substitutions that may simplify the problem. There is an ongoing exploration of different interpretations of the function and its Laplace transform, with no explicit consensus reached.

Contextual Notes

Participants note confusion regarding the limits of integration and the application of the Laplace transform to functions defined over specific intervals. The original poster expresses uncertainty about the evaluation process and the implications of the function's definition.

GreenPrint
Messages
1,186
Reaction score
0

Homework Statement



Find the Laplace transform of

f(t) = t \forall 0≤t≤T, 0 otherwise

Homework Equations





The Attempt at a Solution


I write the function as

tu(t)-t*u(t-T)

That is turn on the function t at t=0 and turn the function t off at t=T. It seems to be right to me.

But now I struggle with trying to take the Laplace transform of this. I know that L(u(t)) = \frac{1}{s}. I know that L(f(t-T)) = e^{-sT}F(s). So I know that L(u(t-T)) = e^{-sT}\frac{1}{s}, but I'm not sure how to evaluate L(t*u(t-T)) because of the extra t term, hence I'm stuck.

Thanks for any help.

I seem to be confused because F(s) = ∫_{0^{-}}^{∞}f(t)e^{-st}dt. So I don't see how you can take the Laplace transform over a domain other than over 0^{-}≤t≤∞, which this question seems to be asking me to do.
 
Last edited:
Physics news on Phys.org
You have a problem of the form ##\mathcal L(f(t)u(t-T))##.$$
\mathcal L(f(t)u(t-T))=\int_0^\infty e^{-st}f(t)u(t-T)~dt =\int_T^\infty e^{-st}f(t)\cdot 1~dt$$Now make the substitution ##u = t - T##:$$
=\int_0^\infty e^{-s(u+T)}f(u+T)~du=e^{-Ts}\int_0^\infty e^{-su}f(u+T)~du
=e^{-Ts}\mathcal Lf(t+T)$$So for your problem$$
\mathcal L(tu(t-T)) = e^{-Ts}\mathcal L(t+T) =~?$$
 
So this problem becomes

L(t*u(t) - t*u(t-T)) = \frac{1}{s^{2}} - L(t*u(t-T)) = \frac{1}{s^{2}} - \int^{∞}_{0^{-1}}e^{-st}t*u(t-T)dt = \frac{1}{s^{2}} - \int^{∞}_{0^{-}}e^{-st}tdt

Now I do u substitution to evaluate the integral. u(t) =t, \frac{du}{dt} = 1. \int dv = \int e^{-st}dt = -\frac{e^{-st}}{s}

Now I continue solving

\frac{1}{s^{2}} - \int^{∞}_{0^{-}}e^{-st}tdt = \frac{1}{s^{2}} -(-\frac{te^{-st}}{s} - \int^{∞}_{0^{-}}e^{-st}dt) = \frac{1}{s^{2}} + \frac{te^{-st}}{s} - \frac{e^{-st}}{s}

The solutions has
F(s) = \frac{1}{s^{2}} - \frac{1}{s^{2}}e^{-sT} - \frac{T}{s}e^{-sT}. So I must be doing something but I don't see what.
 
LCKurtz said:
I gave you a proof that$$
\mathcal L(f(t)u(t-T))=e^{-Ts}\mathcal Lf(t+T)$$

GreenPrint said:
So this problem becomes

L(t*u(t) - t*u(t-T)) = \frac{1}{s^{2}} - L(t*u(t-T)) = \frac{1}{s^{2}} - \int^{∞}_{0^{-1}}e^{-st}t*u(t-T)dt = \frac{1}{s^{2}} - \int^{∞}_{0^{-}}e^{-st}tdt

Now I do u substitution to evaluate the integral. u(t) =t, \frac{du}{dt} = 1. \int dv = \int e^{-st}dt = -\frac{e^{-st}}{s}

Now I continue solving

\frac{1}{s^{2}} - \int^{∞}_{0^{-}}e^{-st}tdt = \frac{1}{s^{2}} -(-\frac{te^{-st}}{s} - \int^{∞}_{0^{-}}e^{-st}dt) = \frac{1}{s^{2}} + \frac{te^{-st}}{s} - \frac{e^{-st}}{s}

The solutions has
F(s) = \frac{1}{s^{2}} - \frac{1}{s^{2}}e^{-sT} - \frac{T}{s}e^{-sT}. So I must be doing something but I don't see what.

You don't have to work out any integrals and re-do all the work. Just use the formula I gave you above. All you have to do is use that formula with ##f(t)=t##.
 
Thanks, I was able to figure it out
 

Similar threads

Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
Replies
3
Views
2K