← Back to Math 055

7.1 Definition of the Laplace Transform: Class Notes

7.1a

The Laplace transform is a tool for solving differential equations.

Definition

A transform transforms one function to another function.

Examples of transforms

The derivative and the integral:

ddx(tan−1⁡x)=11+x2 and ddx(x3)=3x2 are transforms.

∫1xdx=ln⁡|x|+C is a transform.

Transforms possess the linearity properties, e.g.,

ddx[af(x)+bg(x)]=addx[f(x)]+bddx[g(x)]

A constant multiple comes out in front of the transform, and a sum is transformed term by term.

Definition 7.1.1

Let f(t) be a function defined for t≥0. Then the integral ℒ{f(t)}=∫0∞e−stf(t)dt is said to be the Laplace transform of f, provided that the integral converges.

The factor e−st is part of the definition; it is always there. The function f(t) is what is being transformed.

The Laplace transform possesses the linearity properties.

Example 1

Evaluate ℒ{1}.

Solution

ℒ{1}=∫0∞e−st·1dt=−1se−st|0∞=−1sest|0∞=1s,s>0

For s>0, est→∞ as t→∞, so the value at the upper limit is 0. Only the lower limit contributes: 0−(−1s)=1s.

Answer: ℒ{1}=1s,s>0

Example 2

Evaluate ℒ{t}.

Solution

ℒ{t}=∫0∞e−st·tdt

Integrate by parts with u=t,dv=e−stdt, so du=dt,v=−1se−st.

ℒ{t}=−tse−st|0∞+∫0∞1se−stdt=−1s2est|0∞=1s2

The first term is 0: written as tsest, the exponential est grows much faster than t, so the term goes to 0 as t→∞, and it is 0 at t=0. This holds for s>0.

Answer: ℒ{t}=1s2,s>0

Example

Evaluate ℒ{4t−7}

Solution

Using the linearity property, we have

ℒ{4t−7}=4ℒ{t}−7ℒ{1}=4s2−7s

Answer: ℒ{4t−7}=4s2−7s

7.1b

Example

Evaluate ℒ{sin⁡kt}

Solution

ℒ{sin⁡kt}=∫0∞e−stsin⁡ktdt=I

Integrate by parts with u=sin⁡kt,dv=e−stdt, so du=kcos⁡ktdt,v=−1se−st.

I=−1se−stsin⁡kt|0∞+∫0∞kse−stcos⁡ktdt

The first term is 0: as t→∞, e−st→0 (and sin⁡kt stays between −1 and 1), and at t=0, sin⁡0=0. Integrate by parts again:

Integrate by parts with u=kscos⁡kt,dv=e−stdt, so du=−k2ssin⁡ktdt,v=−1se−st.

I=−ks2e−stcos⁡kt|0∞−k2s2∫0∞e−stsin⁡ktdt

At the upper limit the first term is 0; at the lower limit it is −ks2, so the first term is ks2. The integral that remains is I again:

I=ks2−k2s2I(k2s2+1)I=ks2k2+s2s2I=ks2I=ks2+k2

Answer: ℒ{sin⁡kt}=∫0∞e−stsin⁡ktdt=ks2+k2

Theorem 7.1.1: Transforms of Some Basic Functions

7.1c

Definition

A function f is said to be of exponential order if there exist constants c,M>0, and T>0 such that |f(t)|≤Mect for all t>T.

In words: f does not grow faster than an exponential in the long run.

A function f(t) that stays below the exponential M e to the ct for all t greater than T

Graph of a function f(t) (drawn thicker, in blue) and the dashed exponential curve Mect. Near t=0 the graph of f has a bump that rises above the exponential, but for every t>T the graph of f stays below Mect.

Theorem: Sufficient Conditions for Existence

If f is piecewise continuous on [0,∞) and of exponential order, then ℒ{f(t)} exists for s>c.

Proof

f is of exponential order, so there are constants c,M>0,T>0 such that |f(t)|≤Mect for t>T. Split the integral at T:

ℒ{f(t)}=∫0Te−stf(t)dt+∫T∞e−stf(t)dt=I1+I2

I1 exists because f is piecewise continuous (and the interval [0,T] is finite). For I2, use the bound on |f(t)|:

|I2|≤∫T∞e−st|f(t)|dt≤M∫T∞e−stectdt=M∫T∞e−(s−c)tdt=Me−(s−c)t−(s−c)|T∞=0+Ms−ce−(s−c)T,

which is finite if s>c. I1 converges and I2 converges, so I=I1+I2 converges. ✓

“Sufficient” is a guarantee. The conditions given are not necessary conditions. That is, in some cases, f is not of exponential order or not piecewise continuous, but ℒ{f(t)} exists.

Theorem: Behavior of F(s) as s → ∞

If f is piecewise continuous on [0,∞) and of exponential order and F(s)=ℒ{f(t)}, then lims→∞F(s)=0.

Proof

f is of exponential order, so there exist γ,M1>0,T>0 such that |f(t)|≤M1eγt for t>T. (γ is the Greek letter gamma.)

f is piecewise continuous, so f is bounded on 0≤t≤T: |f(t)|≤M2=M2e0t.

Let M=max{M1,M2} and c=max{γ,0}, so |f(t)|≤Mect for all t≥0. Then

|F(s)|≤∫0∞e−st|f(t)|dt≤M∫0∞e−stectdt=Ms−c

for s>c. As s→∞, Ms−c→0, so |F(s)|→0; that is, ℒ{f(t)}=F(s)→0.

7.1d

Example

Use Definition 7.1.1 to find ℒ{f(t)}.

f(t)={2t+1,0≤t<10,t≥1

Solution

Split the integral where the formula changes: on [0, 1) the function is 2t + 1, and for t ≥ 1 it is 0.

ℒ{f(t)}=∫0∞e−stf(t)dt=∫01(2t+1)e−stdt+∫1∞0dt

The second integral is 0. Integrate the first by parts with u=2t+1, dv=e−stdt, so du=2dt, v=−1se−st:

ℒ{f(t)}=−2t+1sest|01+∫012se−stdt=−3ses+1s−2s2est|01=−3ses+1s−2s2es+2s2=1s(1−3e−s)+2s2(1−e−s)

Answer: ℒ{f(t)}=1s(1−3e−s)+2s2(1−e−s)

Example

Use Definition 7.1.1 to find ℒ{f(t)}.

f(t)=e−2t−5

Solution

ℒ{e−2t−5}=∫0∞e−ste−2t−5dt=e−5∫0∞e−(s+2)tdt=e−51−(s+2)e(s+2)t|0∞=e−5(1s+2)=e−5s+2,s>−2

The factor e−5 is a constant, so it comes out of the integral. The value at the upper limit is 0 when s+2>0, that is, s>−2.

Answer: ℒ{e−2t−5}=e−5s+2,s>−2

Example

Use Theorem 7.1.1 to find ℒ{f(t)}.

f(t)=t2−e−9t+5

Solution

By linearity, then parts (b) with n=2, (c) with a=−9, and (a):

ℒ{t2−e−9t+5}=ℒ{t2}−ℒ{e−9t}+5ℒ{1}=2s3−1s+9+5s

Answer: ℒ{t2−e−9t+5}=2s3−1s+9+5s

f(t)=cos2⁡t

Solution

cos2⁡t is not in the table, so first use the identity cos2⁡t=12(1+cos⁡2t):

cos2⁡t=12(1+cos⁡2t)=12+12cos⁡2t

Now use parts (a) and (e) with k=2:

ℒ{cos2⁡t}=12s+s2(s2+4)

Answer: ℒ{cos2⁡t}=12s+s2(s2+4)