Documente Academic
Documente Profesional
Documente Cultură
"ℒ" redirects here. For the Lagrangian, see Lagrangian century by Abel, Lerch, Heaviside, and Bromwich.
mechanics. From 1744, Leonhard Euler investigated integrals of the
form
In mathematics the Laplace transform is an integral
transform named after its discoverer Pierre-Simon ∫ ∫
Laplace (/ləˈplɑːs/). It takes a function of a positive real z = X(x)eax dx and z = X(x)xA dx
variable t (often time) to a function of a complex variable
s (frequency). as solutions of differential equations but did not pursue
[2]
The Laplace transform is very similar to the Fourier the matter very far.
transform. While the Fourier transform of a function Joseph Louis Lagrange was an admirer of Euler and, in
is a complex function of a real variable (frequency), the his work on integrating probability density functions, in-
Laplace transform of a function is a complex function of vestigated expressions of the form
a complex variable. Laplace transforms are usually re-
stricted to functions of t with t > 0. A consequence of ∫
this restriction is that the Laplace transform of a func- X(x)e−ax ax dx,
tion is a holomorphic function of the variable s. Un-
like the Fourier transform, the Laplace transform of a which some modern historians have interpreted within
distribution is generally a well-behaved function. Also modern Laplace transform theory.[3][4]
techniques of complex variables can be used directly to
study Laplace transforms. As a holomorphic function, These types of integrals seem first to have attracted
the Laplace transform has a power series representation. Laplace’s attention in 1782 where he was following in the
This power series expresses a function as a linear super- spirit of Euler in using the integrals themselves as solu-
position of moments of the function. This perspective has tions of equations.[5] However, in 1785, Laplace took the
applications in probability theory. critical step forward when, rather than just looking for a
solution in the form of an integral, he started to apply the
The Laplace transform is invertible on a large class of transforms in the sense that was later to become popular.
functions. The inverse Laplace transform takes a func- He used an integral of the form
tion of a complex variable s (often frequency) and yields a
function of a real variable t (time). Given a simple mathe- ∫
matical or functional description of an input or output to
xs ϕ(x) dx,
a system, the Laplace transform provides an alternative
functional description that often simplifies the process of akin to a Mellin transform, to transform the whole of
analyzing the behavior of the system, or in synthesizing a a difference equation, in order to look for solutions of
new system based on a set of specifications.[1] So, for ex- the transformed equation. He then went on to apply the
ample, Laplace transformation from the time domain to Laplace transform in the same way and started to derive
the frequency domain transforms differential equations some of its properties, beginning to appreciate its poten-
into algebraic equations and convolution into multiplica- tial power.[6]
tion. It has many applications in the sciences and tech-
nology. Laplace also recognised that Joseph Fourier's method of
Fourier series for solving the diffusion equation could
only apply to a limited region of space because those
solutions were periodic. In 1809, Laplace applied his
1 History transform to find solutions that diffused indefinitely in
space.[7]
The Laplace transform is named after mathematician and
astronomer Pierre-Simon Laplace, who used a similar
transform (now called the z-transform) in his work on 2 Formal definition
probability theory. The current widespread use of the
transform (mainly in engineering) came about soon af- The Laplace transform is a frequency-domain approach
ter World War II although it had been used in the 19th for continuous time signals irrespective of whether the
1
2 2 FORMAL DEFINITION
where the lower limit of 0− is shorthand notation for 2.3 Inverse Laplace transform
Main article: Inverse Laplace transform
∫ ∞
lim .
ε↓0 −ε Two integrable functions have the same Laplace trans-
This limit emphasizes that any point mass located at 0 form only if they differ on a set of Lebesgue measure
is entirely captured by the Laplace transform. Although zero. This means that, on the range of the transform,
with the Lebesgue integral, it is not necessary to take such there is an inverse transform. In fact, besides integrable
a limit, it does appear more naturally in connection with functions, the Laplace transform is a one-to-one map-
the Laplace–Stieltjes transform. ping from one function space into another in many other
function spaces as well, although there is usually no easy
characterization of the range. Typical function spaces in
2.1 Probability theory which this is true include the spaces of bounded contin-
uous functions, the space L∞ (0, ∞), or more generally
In pure and applied probability, the Laplace transform tempered functions (that is, functions of at worst polyno-
is defined as an expected value. If X is a random vari- mial growth) on (0, ∞). The Laplace transform is also
able with probability density function f, then the Laplace defined and injective for suitable spaces of tempered dis-
transform of f is given by the expectation tributions.
3
In these cases, the image of the Laplace transform lives in sometimes called the strip of absolute convergence. The
a space of analytic functions in the region of convergence. Laplace transform is analytic in the region of absolute
The inverse Laplace transform is given by the following convergence.
complex integral, which is known by various names (the Similarly, the set of values for which F(s) converges (con-
Bromwich integral, the Fourier–Mellin integral, and ditionally or absolutely) is known as the region of condi-
Mellin’s inverse formula): tional convergence, or simply the region of convergence
(ROC). If the Laplace transform converges (condition-
∫ γ+iT
ally) at s = s0 , then it automatically converges for all s with
1 Re(s) > Re(s0 ). Therefore, the region of convergence is a
f (t) = L−1 {F }(t) = lim est F (s) ds,
2πi T →∞ γ−iT half-plane of the form Re(s) > a, possibly including some
points of the boundary line Re(s) = a.
where γ is a real number so that the contour path of inte-
gration is in the region of convergence of F(s). An alter- In the region of convergence Re(s) > Re(s0 ), the Laplace
native formula for the inverse Laplace transform is given transform of f can be expressed by integrating by parts
by Post’s inversion formula. The limit here is interpreted as the integral
in the weak-* topology.
In practice, it is typically more convenient to decompose ∫ ∞ ∫ u
a Laplace transform into known transforms of functions F (s) = (s−s0 ) e−(s−s0 )t β(t) dt, β(u) = e−s0 t f (t) dt.
obtained from a table, and construct the inverse by in- 0 0
∫ ∞
f (t)e−st dt
0 4 Properties and theorems
exists (as a proper Lebesgue integral). The Laplace trans-
The Laplace transform has a number of properties that
form is usually understood as conditionally convergent,
meaning that it converges in the former instead of the lat- make it useful for analyzing linear dynamical systems.
ter sense. The most significant advantage is that differentiation and
integration become multiplication and division, respec-
The set of values for which F(s) converges absolutely is tively, by s (similarly to logarithms changing multiplica-
either of the form Re(s) > a or else Re(s) ≥ a, where a tion of numbers to addition of their logarithms).
is an extended real constant, −∞ ≤ a ≤ ∞. (This follows
from the dominated convergence theorem.) The constant Because of this property, the Laplace variable s is also
a is known as the abscissa of absolute convergence, and known as operator variable −1
in the L domain: either
[10]
depends on the growth behavior of f(t). Analogously, derivative operator or (for s ) integration operator. The
the two-sided transform converges absolutely in a strip transform turns integral equations and differential equa-
of the form a < Re(s) < b, and possibly including the tions to polynomial equations, which are much easier to
lines Re(s) = a or Re(s) = b. [11]
The subset of values of solve. Once solved, use of the inverse Laplace transform
s for which the Laplace transform converges absolutely reverts to the time domain.
is called the region of absolute convergence or the do- Given the functions f(t) and g(t), and their respective
main of absolute convergence. In the two-sided case, it is Laplace transforms F(s) and G(s),
4 4 PROPERTIES AND THEOREMS
∫ ∫ ∞
∞
f (t)x dtt L {f (t)} = e−st f (t) dt
0−
0 [ ]∞ ∫ ∞ −st
f (t)e−st e
where the discrete function a(n) is replaced by the con- = − f ′ (t) dt parts) (by
−s 0− 0− −s
tinuous one f(t). (See Mellin transform below.) [ ]
f (0− ) 1
Changing the base of the power from x to e gives = − + L {f ′ (t)} ,
−s s
∫ ∞ yielding
( )t
f (t) eln x dt
0
For this to converge for, say, all bounded functions f, it is L {f ′ (t)} = s · L {f (t)} − f (0− ),
necessary to require that ln x < 0. Making the substitution
−s = ln x gives just the Laplace transform: and in the bilateral case,
∫ ∞ ∫ ∞
′
f (t)e −st
dt L {f (t)} = s e−st f (t) dt = s · L{f (t)}.
0 −∞
4.5 Relationship to other transforms 5
The general result then the Laplace–Stieltjes transform of g and the Laplace
transform of f coincide. In general, the Laplace–Stieltjes
{ } transform is the Laplace transform of the Stieltjes mea-
− (n−1) −sure associated to g. So in practice, the only distinc-
L f (t) = s ·L {f (t)}−s
(n) n n−1
f (0 )−· · ·−f (0 ),
tion between the two transforms is that the Laplace trans-
where f (n) denotes the nth derivative of f, can then be form is thought of as operating on the density function
established with an inductive argument. of the measure, whereas the Laplace–Stieltjes transform
is thought of as operating on its cumulative distribution
function.[14]
4.4 Evaluating improper integrals
4.5.2 Fourier transform
Let L {f (t)} = F (s) , then (see the table above)
The continuous Fourier transform is equivalent to evalu-
{ } ∫ ∞ ating the bilateral Laplace transform with imaginary ar-
f (t)
L = F (p) dp, gument s = iω or s = 2πfi,[15]
t s
or
fˆ(ω) = F{f (t)}
= L{f (t)}|s=iω = F (s)|s=iω
∫ ∞ ∫ ∞ ∫ ∞
f (t) −st
t
e dt = F (p) dp. = e−iωt f (t) dt .
0 s −∞
Letting s → 0, gives one the identity This definition of the Fourier transform requires a pref-
actor of 1/2 π on the reverse Fourier transform. This
relationship between the Laplace and Fourier transforms
∫ ∞ ∫ ∞
f (t) is often used to determine the frequency spectrum of a
dt = F (p) dp.
0 t 0
signal or dynamical system.
provided that the interchange of limits can be justified. The above relation is valid as stated if and only if the re-
Even when the interchange cannot be justified the calcu- gion of convergence (ROC) of F(s) contains the imagi-
lation can be suggestive. For example, proceeding for- nary axis, σ = 0.
mally one has For example, the function f(t) = cos(ω0 t) has a Laplace
transform F(s) = s/(s2 + ω0 2 ) whose ROC is Re(s) > 0.
∫ ∫ ∞( As)s = iω is a pole of F(s),
∞ substituting s = iω in F(s)
∞
1 p p does not yield
1 the + a2 transform of f(t)u(t), which is
p2Fourier
(cos(at) − cos(bt)) dt = − 2 dp = ln 2 = ln b−ln a.
0 t 0 p2 + a2 p + bproportional
2 2 to the
p + b2 0delta-function δ(ω − ω0 ).
Dirac
The validity of this identity can be proved by other means. However, a relation of the form
It is an example of a Frullani integral.
Another example is Dirichlet integral. lim F (σ + iω) = fˆ(ω)
σ→0+
∫ x ∫ ∞
dθ
g(x) = f (t) dt G(s) = M{g(θ)} = θs g(θ)
0 0 θ
6 5 TABLE OF SELECTED LAPLACE TRANSFORMS
we set θ = e−t we get a two-sided Laplace transform. Comparing the last two equations, we find the relation-
ship between the unilateral Z-transform and the Laplace
transform of the sampled signal,
4.5.4 Z-transform
where T = 1/fs is the sampling period (in units of time 4.5.5 Borel transform
e.g., seconds) and fs is the sampling rate (in samples per
second or hertz). The integral form of the Borel transform
Let
∫ ∞
F (s) = f (z)e−sz dz
∞
∑ 0
def
∆T (t) = δ(t − nT )
n=0
is a special case of the Laplace transform for f an entire
function of exponential type, meaning that
be a sampling impulse train (also called a Dirac comb)
and
|f (z)| ≤ AeB|z|
∞
∑
def
xq (t) = x(t)∆T (t) = x(t) δ(t − nT ) for some constants A and B. The generalized Borel trans-
n=0
form allows a different weighting function to be used,
∞
∑ ∞
∑ rather than the exponential function, to transform func-
= x(nT )δ(t − nT ) = x[n]δ(t − nT ) tions not of exponential type. Nachbin’s theorem gives
n=0 n=0 necessary and sufficient conditions for the Borel trans-
form to be well defined.
be the sampled representation of the continuous-time x(t)
dN
= −λN,
dt
where λ is the decay constant. The Laplace transform can
be used to solve this equation.
Rearranging the equation to one side, we have
dN
+ λN = 0.
dt
Next, we take the Laplace transform of both sides of the
s-domain equivalent circuits
equation:
Note that the resistor is exactly the same in the time do- ( )
main and the s-domain. The sources are put in if there are sÑ (s) − N0 + λÑ (s) = 0,
8 7 EXAMPLES: HOW TO APPLY THE PROPERTIES AND THEOREMS
Solving, we find
V (s)
Z(s) = .
I(s) V0 =0
N0
Ñ (s) = . Using this definition and the previous equation, we find:
s+λ
Finally, we take the inverse Laplace transform to find the
general solution 1
Z(s) = ,
sC
{ }
−1 −1 N0 which is the correct expression for the complex
N (t) = L {Ñ (s)} = L impedance of a capacitor.
s+λ
= N0 e−λt ,
which is indeed the correct form for radioactive decay. 7.3 Example 3: Method of partial fraction
expansion
7.2 Example 2: Deriving the complex Consider a linear time-invariant system with transfer
impedance for a capacitor function
I(s) = C(sV (s) − V0 ), The unknown constants P and R are the residues lo-
cated at the corresponding poles of the transfer function.
where Each residue represents the relative contribution of that
singularity to the transfer function’s overall shape.
By the residue theorem, the inverse Laplace transform
I(s) = L{i(t)}, depends only upon the poles and their residues. To find
V (s) = L{v(t)}, the residue P, we multiply both sides of the equation by s
+ α to get
and
1 R(s + α)
V0 = v(t)|t=0 . =P+ .
s+β s+β
7.6 Example 5: Phase delay 9
Note that
{ }
s β−α
x(t) = e−αt L−1 + 2
s2 + ω 2 s + ω2
−1
R= = −P { ( )( )}
β−α s β−α ω
= e−αt L−1 +
s2 + ω 2 ω s2 + ω 2
and so the substitution of R and P into the expanded ex-
pression for H(s) gives [ { } ( ) { }]
s β−α ω
= e−αt L−1 + L−1 .
s2 + ω 2 ω s2 + ω 2
( ) ( )
1 1 1 Finally, using the Laplace transforms for sine and cosine
H(s) = · − .
β−α s+α s+β (see the table, above), we have
Finally, using the linearity property and the known trans-
form for exponential decay (see Item #3 in the Table [ ( ) ]
−αt β−α
of Laplace Transforms, above), we can take the inverse x(t) = e cos (ωt)u(t) + sin (ωt)u(t) .
ω
Laplace transform of H(s) to obtain [ ( ) ]
−αt β−α
x(t) = e cos (ωt) + sin (ωt) u(t).
ω
1 ( −αt )
h(t) = L−1 {H(s)} = e − e−βt ,
β−α
7.6 Example 5: Phase delay
which is the impulse response of the system.
Starting with the Laplace transform,
1 1 1 s sin(ϕ) ω cos(ϕ)
H(s) = = · X(s) = + 2
s2 + ω 2 s + ω2
(s + a)(s + b) s+a s+b ( ) ( )
s ω
is = sin(ϕ) 2 + cos(ϕ) 2 .
s + ω2 s + ω2
Starting with the Laplace transform This is just the sine of the sum of the arguments, yielding:
10 9 NOTES
9 Notes
x(t) = sin(ωt + ϕ).
[1] Korn & Korn 1967, §8.1
We can apply similar logic to find that
[2] Euler 1744, Euler 1753, Euler 1769
{ }
s cos ϕ − ω sin ϕ
L−1 = cos (ωt + ϕ). [3] Lagrange 1773
s2 + ω 2
The wide and general applicability of the Laplace trans- [6] Grattan-Guinness 1997, pp. 261–262
form and its inverse is illustrated by an application in as-
tronomy which provides some information on the spa- [7] Grattan-Guinness 1997, pp. 262–266
tial distribution of matter of an astronomical source of
radiofrequency thermal radiation too distant to resolve [8] Feller 1971, §XIII.1
as more than a point, given its flux density spectrum,
rather than relating the time domain with the spectrum [9] The cumulative distribution function is the integral of the
(frequency domain). probability density function.
Assuming certain properties of the object, e.g. spher-
ical shape and constant temperature, calculations based [10] Widder 1941, Chapter II, §1
on carrying out an inverse Laplace transformation on
the spectrum of the object can produce the only possi- [11] Widder 1941, Chapter VI, §2
ble model of the distribution of matter in it (density as
a function of distance from the center) consistent with [12] Korn & Korn 1967, pp. 226–227
the spectrum.[20] When independent information on the
structure of an object is available, the inverse Laplace [13] Bracewell 2000, Table 14.1, p. 385
transform method has been found to be in good agree-
ment. [14] Feller 1971, p. 432
13.2 Images
• File:Commons-logo.svg Source: https://upload.wikimedia.org/wikipedia/en/4/4a/Commons-logo.svg License: CC-BY-SA-3.0 Contribu-
tors: ? Original artist: ?
• File:S-Domain_circuit_equivalents.svg Source: https://upload.wikimedia.org/wikipedia/commons/2/22/S-Domain_circuit_
equivalents.svg License: CC BY-SA 4.0 Contributors: Own work Original artist: Krishnavedala