Laplace Transform: Difference between revisions

From Class Wiki
Jump to navigation Jump to search
No edit summary
 
(68 intermediate revisions by 6 users not shown)
Line 1: Line 1:
Standard Form:
Laplace transforms are an adapted integral form of a differential equation (created and introduced by the French mathematician Pierre-Simon Laplace (1749-1827)) used to describe electrical circuits and physical processes. Adapted from previous notions given by other notable mathematicians and engineers like Joseph-Louis Lagrange (1736-1812) and Leonhard Euler (1707-1783), Laplace transforms are used to be a more efficient and easy-to-recognize form of a mathematical equation.
 
==Standard Form==
This is the standard form of a Laplace transform that a function will undergo.
:<math>F(s) = \mathcal{L} \left\{f(t)\right\}=\int_0^{\infty} e^{-st} f(t) \,dt </math>
:<math>F(s) = \mathcal{L} \left\{f(t)\right\}=\int_0^{\infty} e^{-st} f(t) \,dt </math>


Sample Functions:
== Sample Functions ==
:<math>F(s) = \mathcal{L} \left\{1\right\}=\int_0^{\infty} e^{-st} \,dt = </math> <math> \frac {1}{s}</math>
 
 
The following is a list of commonly seen functions of which the Laplace transform is taken. The start function is noted within the Laplace symbol <math> \mathcal{L} \left\{ \right\} </math>.
 
: <math>F(s) = \mathcal{L} \left\{1\right\}=\int_0^{\infty} e^{-st} \,dt = </math> <math> \frac {1}{s}</math>
 
: <math>F(s) = \mathcal{L} \left\{t^n\right\}=\int_0^{\infty} e^{-st} t^n \,dt = </math> <math> \frac {n!}{s^{n+1}}</math>
 
: <math>F(s) = \mathcal{L} \left\{e^{at}\right\}=\int_0^{\infty} e^{-st} e^{at} \,dt = </math> <math> \frac {1}{s-a}</math>
 
: <math>F(s) = \mathcal{L} \left\{sin(\omega t)\right\}=\int_0^{\infty} e^{-st} sin(\omega t) \,dt = </math> <math> \frac {\omega}{s^2+\omega^2}</math>
 
: <math>F(s) = \mathcal{L} \left\{cos(\omega t)\right\}=\int_0^{\infty} e^{-st} cos(\omega t) \,dt = </math> <math> \frac {s}{s^2+\omega^2}</math>
 
: <math>F(s) = \mathcal{L} \left\{t^n g(t)\right\}=\int_0^{\infty} e^{-st} t^n g(t) \,dt = </math> <math> \frac {(-1)^n d^n G(s)} {ds^n}  \mbox{ for}~n\ \mbox{= 1,2,...}</math>
 
: <math>F(s) = \mathcal{L} \left\{t sin(\omega t)\right\}=\int_0^{\infty} e^{-st} t sin(\omega t) \,dt = </math> <math> \frac {2 \omega s} {(s^2+\omega^2)^2} </math>
 
: <math>F(s) = \mathcal{L} \left\{t cos(\omega t)\right\}=\int_0^{\infty} e^{-st} t cos(\omega t) \,dt = </math> <math> \frac {s^2-\omega^2} {(s^2+\omega^2)^2} </math>
 
: <math>F(s) = \mathcal{L} \left\{g(at)\right\}=\int_0^{\infty} e^{-st} g(at) \,dt = </math> <math> \frac {1} {a} G \left(\frac {s} {a}\right), \mbox{ for} ~a > 0.</math>
 
: <math>F(s) = \mathcal{L} \left\{e^{at} g(t)\right\}=\int_0^{\infty} e^{-st} e^{at} g(t) \,dt = G(s-a) </math>
 
: <math>F(s) = \mathcal{L} \left\{e^{at} t^n\right\}=\int_0^{\infty} e^{-st} e^{at} t^n \,dt = </math> <math> \frac {n!} {(s-a)^{n+1}} \mbox{ for}~n\ \mbox{= 1,2,...}</math>
 
: <math>F(s) = \mathcal{L} \left\{te^{-t}\right\}=\int_0^{\infty} e^{-st} te^{-t} \,dt = </math> <math> \frac {1} {(s+1)^2} </math>
 
: <math>F(s) = \mathcal{L} \left\{1-e^{-t/T}\right\}=\int_0^{\infty} e^{-st} (1-e^{-t/T}) \,dt = </math> <math> \frac {1} {s(1+Ts)} </math>
 
: <math>F(s) = \mathcal{L} \left\{e^{at} sin(\omega t)\right\}=\int_0^{\infty} e^{-st} e^{at} sin(\omega t) \,dt = </math> <math> \frac {\omega} {(s-a)^2 + \omega^2} </math>
 
: <math>F(s) = \mathcal{L} \left\{e^{at} cos(\omega t)\right\}=\int_0^{\infty} e^{-st} e^{at} cos(\omega t) \,dt = </math> <math> \frac {s-a} {(s-a)^2 + \omega^2} </math>
 
: <math>F(s) = \mathcal{L} \left\{u(t)\right\}=\int_0^{\infty} e^{-st} u(t) \,dt = </math> <math> \frac {1} {s} </math>
 
: <math>F(s) = \mathcal{L} \left\{u(t-a)\right\}=\int_0^{\infty} e^{-st} u(t-a) \,dt = </math> <math> \frac {e^{-as}} {s} </math>
 
: <math>F(s) = \mathcal{L} \left\{u(t-a) g(t-a)\right\}=\int_0^{\infty} e^{-st} u(t-a) g(t-a) \,dt = e^{-as} G(s) </math>
 
: <math>F(s) = \mathcal{L} \left\{g'(t)\right\}=\int_0^{\infty} e^{-st} g'(t) \,dt = sG(s) - g(0) </math>
 
: <math>F(s) = \mathcal{L} \left\{g''(t)\right\}=\int_0^{\infty} e^{-st} g''(t) \,dt = s^2 \cdot G(s) - s \cdot g(0) - g'(0) </math>
 
: <math>F(s) = \mathcal{L} \left\{g^{(n)}(t)\right\}=\int_0^{\infty} e^{-st} g^{(n)}(t) \,dt = s^n \cdot G(s) - s^{n-1} \cdot g(0) - s^{n-2} \cdot g'(0) - ... - g^{(n-1)}(0) </math>
 
==Transfer Function==
The Laplace transform of the impulse response of a circuit with no initial conditions is called the transfer function. If a single-input, single-output circuit has no internal stored energy and all the independent internal sources are zero, the transfer function is
 
: <math>H(s) = \frac{ \mathcal{L} (response~signal)} { \mathcal{L} (input~signal)}. </math>
 
Impedances and admittances are special cases of transfer functions.
 
==Example==
Solve the differential equation:
:<math>y''-2y'-15y=6  \qquad y(0)=1 \qquad y'(0)=3</math>
 
We start by taking the Laplace transform of each term.
:<math>\mathcal{L} \left\{y''\right\}-2\mathcal{L} \left\{y'\right\}-15\mathcal{L} \left\{y\right\}=\mathcal{L} \left\{6\right\}</math>
 
The next step is to perform the respective Laplace transforms, using the information given above.
:<math> (s^2\mathcal{L} \left\{y\right\}-s-3)-2(s\mathcal{L} \left\{y\right\}-1)-15\mathcal{L} \left\{y\right\} = \mathcal{L} \left\{6\right\}</math>
 
Using association, the equation is rearranged:
:<math> (s^2-2s-15)\mathcal{L} \left\{y\right\}= \frac {6} {s} +s+3-2</math>
 
Continuing on using the method of partial fractions, the equation is progressed:
:<math> \frac {s^2+s+6} {s(s+3)(s-5)} = \frac {A} {s} + \frac {B} {s+3} + \frac {C} {s-5}</math>
 
 
:<math> A(s+3)(s-5)+Bs(s-5)+Cs(s+3)=s^2+s+6 \,</math>
 
 
:<math> A+B+C=-1 \,</math>
:<math> -2A-5B+3C=1 \,</math>
:<math> -15A=6 \,</math>
 
 
:<math> A=\frac {-2} {5} \qquad B=\frac {1} {2} \qquad C=\frac {9} {10} </math>
 
Plugging the above values back into the equation further up, we get:
:<math> \mathcal{L} \left\{y\right\} = \frac {\frac {-2} {5}} {s} + \frac {\frac {1} {2}} {s+3} + \frac {\frac {9} {10}} {s-5} </math>
 
Applying anti-Laplace transforms, we get the equation:
:<math> y= \mathcal{L}^{-1} \left\{\frac {\frac {-2} {5}} {s}\right\}  + \mathcal{L}^{-1} \left\{\frac {\frac {1} {2}} {s+3}\right\}  + \mathcal{L}^{-1} \left\{\frac {\frac {9} {10}} {s-5}\right\}  </math>
 
Applying the Laplace transforms in reverse (as the above equation utilizes inverse Laplace transforms) for the above equation, we get the solution:
:<math> y(t)= \frac {-2} {5} + \frac {1} {2} e^{-3t} + \frac {9} {10} e^{5t} </math>
 
==References==
DeCarlo, Raymond A.; Lin, Pen-Min (2001), Linear Circuit Analysis, Oxford University Press, ISBN 0-19-513666-7 .
 
Zill, Dennis G. (2005), A First Course in Differential Equations with Modeling Applications Ninth Edition, Brooks/Cole Cengage Learning, ISBN 0-495-10824-3 .
 
== External links ==
*[http://www.intmath.com/Laplace-transformation/Intro.php The Laplace Transform].
 
==Authors==
Colby Fullerton
 
Brian Roath
 
==Reviewed By==
David Robbins
 
Thomas Wooley
 
==Read By==
 
 
Jaymin Joseph
<br />
 
John Hawkins
 
==Comments==
John Hawkins:


:<math>F(s) = \mathcal{L} \left\{t^n\right\}=\int_0^{\infty} e^{-st} t^n \,dt = </math> <math> \frac {n!}{s^{n+1}}</math>
*Nice list of transforms!  Where did you find it?  Inside the back cover of the textbook has a good list, but none including transforms of <math>\ g(t)</math>.  I see your reference to the textbook. What page?


:<math>F(s) = \mathcal{L} \left\{e^{at}\right\}=\int_0^{\infty} e^{-st} e^{at} \,dt = </math> <math> \frac {1}{s-a}</math>
** Oops! It seems I forgot to add a reference to the list, thanks for mentioning it. I got the others from a list in the back of the ODE textbook. I'll list the reference up above.


:<math>F(s) = \mathcal{L} \left\{sin(\omega t)\right\}=\int_0^{\infty} e^{-st} sin(\omega t) \,dt = </math> <math> \frac {\omega}{s^2+\omega^2}</math>


:<math>F(s) = \mathcal{L} \left\{cos(\omega t)\right\}=\int_0^{\infty} e^{-st} cos(\omega t) \,dt = </math> <math> \frac {s}{s^2+\omega^2}</math>
*I believe that


:<math>F(s) = \mathcal{L} \left\{t^n g(t)\right\}=\int_0^{\infty} e^{-st} t^n g(t) \,dt = </math> <math> \frac {(-1)^n d^n G(s)} {ds^n} </math> for n=1,2,...
:<math> \frac {6+s-s^2} {s(s+3)(s-5)} = \left(\frac {A} {s}\right) \left(\frac {B} {s+3}\right) \left(\frac {C} {s-5}\right)</math>


:<math>F(s) = \mathcal{L} \left\{t sin(\omega t)\right\}=\int_0^{\infty} e^{-st} t sin(\omega t) \,dt = </math> <math> \frac {2 \omega s} {(s^2+\omega^2)^2} </math>
should be


:<math>F(s) = \mathcal{L} \left\{t cos(\omega t)\right\}=\int_0^{\infty} e^{-st} t cos(\omega t) \,dt = </math> <math> \frac {s^2-\omega^2} {(s^2+\omega^2)^2} </math>
:<math> \frac {6+s-s^2} {s(s+3)(s-5)} = \frac {A} {s}+ \frac {B} {s+3}+ \frac {C} {s-5}</math>


:<math>F(s) = \mathcal{L} \left\{g(t)\right\}=\int_0^{\infty} e^{-st} g(t) \,dt = </math> <math> \frac {1} {a} G \left(\frac {s} {a}\right)</math>
** Thank you for pointing that out. It should be fixed now.


:<math>F(s) = \mathcal{L} \left\{e^{at} g(t)\right\}=\int_0^{\infty} e^{-st} e^{at} g(t) \,dt = </math> <math> G(s-a) </math>
*Also, the solution <math> y(t)= -\frac {2} {5} - \frac {1} {4} e^{-3t} - \frac {7} {20} e^{5t} </math> does not match the initial condition of <math>\ y(0)=1</math>.


:<math>F(s) = \mathcal{L} \left\{e^{at} t^n\right\}=\int_0^{\infty} e^{-st} e^{at} t^n \,dt = </math> <math> \frac {n!} {(s-a)^{n+1}} </math> for n=1,2,...
** Yeah, you're right. A minor sign error was what messed it up. It should be fixed now.

Latest revision as of 22:46, 3 February 2010

Laplace transforms are an adapted integral form of a differential equation (created and introduced by the French mathematician Pierre-Simon Laplace (1749-1827)) used to describe electrical circuits and physical processes. Adapted from previous notions given by other notable mathematicians and engineers like Joseph-Louis Lagrange (1736-1812) and Leonhard Euler (1707-1783), Laplace transforms are used to be a more efficient and easy-to-recognize form of a mathematical equation.

Standard Form

This is the standard form of a Laplace transform that a function will undergo.

F(s)={f(t)}=0estf(t)dt

Sample Functions

The following is a list of commonly seen functions of which the Laplace transform is taken. The start function is noted within the Laplace symbol {}.

F(s)={1}=0estdt= 1s
F(s)={tn}=0esttndt= n!sn+1
F(s)={eat}=0esteatdt= 1sa
F(s)={sin(ωt)}=0estsin(ωt)dt= ωs2+ω2
F(s)={cos(ωt)}=0estcos(ωt)dt= ss2+ω2
F(s)={tng(t)}=0esttng(t)dt= (1)ndnG(s)dsn forn= 1,2,...
F(s)={tsin(ωt)}=0esttsin(ωt)dt= 2ωs(s2+ω2)2
F(s)={tcos(ωt)}=0esttcos(ωt)dt= s2ω2(s2+ω2)2
F(s)={g(at)}=0estg(at)dt= 1aG(sa), fora>0.
F(s)={eatg(t)}=0esteatg(t)dt=G(sa)
F(s)={eattn}=0esteattndt= n!(sa)n+1 forn= 1,2,...
F(s)={tet}=0esttetdt= 1(s+1)2
F(s)={1et/T}=0est(1et/T)dt= 1s(1+Ts)
F(s)={eatsin(ωt)}=0esteatsin(ωt)dt= ω(sa)2+ω2
F(s)={eatcos(ωt)}=0esteatcos(ωt)dt= sa(sa)2+ω2
F(s)={u(t)}=0estu(t)dt= 1s
F(s)={u(ta)}=0estu(ta)dt= eass
F(s)={u(ta)g(ta)}=0estu(ta)g(ta)dt=easG(s)
F(s)={g(t)}=0estg(t)dt=sG(s)g(0)
F(s)={g(t)}=0estg(t)dt=s2G(s)sg(0)g(0)
F(s)={g(n)(t)}=0estg(n)(t)dt=snG(s)sn1g(0)sn2g(0)...g(n1)(0)

Transfer Function

The Laplace transform of the impulse response of a circuit with no initial conditions is called the transfer function. If a single-input, single-output circuit has no internal stored energy and all the independent internal sources are zero, the transfer function is

H(s)=(responsesignal)(inputsignal).

Impedances and admittances are special cases of transfer functions.

Example

Solve the differential equation:

y2y15y=6y(0)=1y(0)=3

We start by taking the Laplace transform of each term.

{y}2{y}15{y}={6}

The next step is to perform the respective Laplace transforms, using the information given above.

(s2{y}s3)2(s{y}1)15{y}={6}

Using association, the equation is rearranged:

(s22s15){y}=6s+s+32

Continuing on using the method of partial fractions, the equation is progressed:

s2+s+6s(s+3)(s5)=As+Bs+3+Cs5


A(s+3)(s5)+Bs(s5)+Cs(s+3)=s2+s+6


A+B+C=1
2A5B+3C=1
15A=6


A=25B=12C=910

Plugging the above values back into the equation further up, we get:

{y}=25s+12s+3+910s5

Applying anti-Laplace transforms, we get the equation:

y=1{25s}+1{12s+3}+1{910s5}

Applying the Laplace transforms in reverse (as the above equation utilizes inverse Laplace transforms) for the above equation, we get the solution:

y(t)=25+12e3t+910e5t

References

DeCarlo, Raymond A.; Lin, Pen-Min (2001), Linear Circuit Analysis, Oxford University Press, ISBN 0-19-513666-7 .

Zill, Dennis G. (2005), A First Course in Differential Equations with Modeling Applications Ninth Edition, Brooks/Cole Cengage Learning, ISBN 0-495-10824-3 .

External links

Authors

Colby Fullerton

Brian Roath

Reviewed By

David Robbins

Thomas Wooley

Read By

Jaymin Joseph

John Hawkins

Comments

John Hawkins:

  • Nice list of transforms! Where did you find it? Inside the back cover of the textbook has a good list, but none including transforms of g(t). I see your reference to the textbook. What page?
    • Oops! It seems I forgot to add a reference to the list, thanks for mentioning it. I got the others from a list in the back of the ODE textbook. I'll list the reference up above.


  • I believe that
6+ss2s(s+3)(s5)=(As)(Bs+3)(Cs5)

should be

6+ss2s(s+3)(s5)=As+Bs+3+Cs5
    • Thank you for pointing that out. It should be fixed now.
  • Also, the solution y(t)=2514e3t720e5t does not match the initial condition of y(0)=1.
    • Yeah, you're right. A minor sign error was what messed it up. It should be fixed now.