文章目录
Laplace 变换
2.1 Laplace变换概念
2.1.1 Laplace变换引入
F o u r i e r Fourier Fourier变换有非常广泛的应用，也有明显的缺点，即对函数 f ( x ) f(x) f(x)的要求太苛刻，这表现在两个方面:
- 当函数在区间 ( − ∞ , ∞ ) (-∞,\infty) (−∞,∞)绝对可积，即满足 ∫ − ∞ ∞ ∣ f ( x ) ∣ d x < ∞ \int_{-\infty}^{\infty}|f(x)|\,{\rm d}x<∞ ∫−∞∞∣f(x)∣dx<∞时，傅里叶变换存在。这个条件要求当 ∣ x ∣ → ∞ |x|→∞ ∣x∣→∞时， f ( x ) → 0 f(x)→0 f(x)→0。事实上，许多函数都不满足这个条件，如 f ( x ) = a ( 常 数 ) f(x)=a(常数) f(x)=a(常数)、正弦和余弦函数、线性函数、单位阶跃函数等。
- 要求函数
f
(
x
)
f(x)
f(x)必须在整个区间
f
(
x
)
f(x)
f(x)有定义，对于定义在区间
0
≤
x
<
∞
0≤x<\infty
0≤x<∞的函数，如以时间
t
t
t为变量的函数
f
(
t
)
f(t)
f(t)，则无法进行
F
o
u
r
i
e
r
Fourier
Fourier变换。
解决这些问题的办法是引入 L a p l a c e Laplace Laplace变换。
2.1.2 Laplace变换的定义
L a p l a c e Laplace Laplace变换是在 F o u r i e r Fourier Fourier变换的基础上引入的。现在考虑对一个任意函数 g ( t ) ( t ≥ 0 ) g(t)(t≥0) g(t)(t≥0)进行 F o u r i e r Fourier Fourier变换，为了使之在 ( − ∞ , ∞ ) (-\infty,\infty) (−∞,∞)区间有定义，给它乘以单位阶跃函数 u ( t ) u(t) u(t)；为了容易满足绝对可积条件，再乘以衰减因子 e x p ( − β t ) ( β > 0 ) exp(-\beta t)(\beta>0) exp(−βt)(β>0)，然后对函数 g ( t ) u ( t ) e x p ( − β t ) g(t)u(t)exp(-\beta t) g(t)u(t)exp(−βt)进行 F o u r i e r Fourier Fourier变换
G
β
(
ω
)
=
∫
−
∞
+
∞
φ
(
t
)
u
(
t
)
e
−
β
t
e
−
i
ω
t
d
t
=
∫
0
+
∞
f
(
t
)
e
−
(
β
+
i
ω
)
t
d
t
=
∫
0
+
∞
f
(
t
)
e
−
s
t
d
t
\begin{aligned}G_{\beta}(\omega)&=\int_{-\infty}^{+\infty}{\varphi(t)u(t)e^{-\beta t}e^{-i\omega t}}\,{\rm d}t\\&=\int_{0}^{+\infty}{f(t)e^{-(\beta+i\omega)t}}\,{\rm d}t=\int_{0}^{+\infty}{f(t)e^{-st}}\,{\rm d}t\end{aligned}
Gβ(ω)=∫−∞+∞φ(t)u(t)e−βte−iωtdt=∫0+∞f(t)e−(β+iω)tdt=∫0+∞f(t)e−stdt
其中
s
=
β
+
i
ω
,
f
(
t
)
=
φ
(
t
)
u
(
t
)
s=\beta+i\omega,f(t)=\varphi(t)u(t)
s=β+iω,f(t)=φ(t)u(t)
若再设
F
(
s
)
=
G
β
(
s
−
β
i
)
F(s)=G_{\beta}\left(\frac{s-\beta}{i}\right)
F(s)=Gβ(is−β)
则得
F
(
s
)
=
∫
0
+
∞
f
(
t
)
e
−
s
t
d
t
F(s)=\int_{0}^{+\infty}{f(t)e^{-st}}\,{\rm d}t
F(s)=∫0+∞f(t)e−stdt
定义
设函数
f
(
t
)
f(t)
f(t)当
t
≥
0
t\ge0
t≥0有定义，而且积分
∫
0
+
∞
f
(
t
)
e
−
s
t
d
t
\int_{0}^{+\infty}{f(t)e^{-st}}\,{\rm d}t
∫0+∞f(t)e−stdt
在
s
s
s的某一域内收敛，则由此积分所确定的函数可写为
F
(
s
)
=
∫
0
+
∞
f
(
t
)
e
−
s
t
d
t
(2.1)
F(s)=\int_{0}^{+\infty}{f(t)e^{-st}}\,{\rm d}t\tag{2.1}
F(s)=∫0+∞f(t)e−stdt(2.1)
我们称
(
2.1
)
(2.1)
(2.1)式为函数
f
(
t
)
f(t)
f(t)的
L
a
p
l
a
c
e
Laplace
Laplace变换式，记为
F
(
s
)
=
L
[
f
(
t
)
]
F(s)=\mathscr{L}\left[f(t)\right]
F(s)=L[f(t)]
F
(
s
)
F(s)
F(s)称为
f
(
t
)
f(t)
f(t)的
L
a
p
l
a
c
e
Laplace
Laplace变换
(
或
称
为
象
函
数
)
(或称为象函数)
(或称为象函数)
若
F
(
s
)
F(s)
F(s)是
f
(
t
)
f(t)
f(t)的
L
a
p
l
a
c
e
Laplace
Laplace变换，则称
f
(
t
)
f(t)
f(t)为
F
(
s
)
F(s)
F(s)的
L
a
p
l
a
c
e
Laplace
Laplace逆变换
(
或
称
为
象
原
函
数
)
(或称为象原函数)
(或称为象原函数)，记为
f
(
t
)
=
L
−
1
[
F
(
s
)
]
f(t)=\mathscr{L}^{-1}\left[F(s)\right]
f(t)=L−1[F(s)]
实际上，
f
(
t
)
(
t
≥
0
)
f(t)(t\ge0)
f(t)(t≥0)的
L
a
p
l
a
c
e
Laplace
Laplace变换，实际上就是
f
(
t
)
u
(
t
)
e
−
β
t
f(t)u(t)e^{-\beta t}
f(t)u(t)e−βt的
F
o
u
r
i
e
r
Fourier
Fourier变换
2.1.3 Laplace变换的存在性定理
若函数 f ( t ) f(t) f(t)满足下列条件：
- 在 t ≥ 0 t\ge0 t≥0的任一有限区间上分段连续；
- 当
t
→
+
∞
t \to +\infty
t→+∞时，
f
(
t
)
f(t)
f(t)的增长速度不超过某一指数函数，亦即存在常数
M
>
0
M>0
M>0及
c
≥
0
c\ge0
c≥0，使得
∣ f ( t ) ∣ ≤ M e c t , 0 ≤ t < + ∞ |f(t)|\le Me^{ct},0\le t<+\infty ∣f(t)∣≤Mect,0≤t<+∞
成立
则
f
(
t
)
f(t)
f(t)的
L
a
p
l
a
c
e
Laplace
Laplace变换
F
(
s
)
=
∫
0
+
∞
f
(
t
)
e
−
s
t
d
t
(2.1)
F(s)=\int_{0}^{+\infty}{f(t)e^{-st}}\,{\rm d}t \tag{2.1}
F(s)=∫0+∞f(t)e−stdt(2.1)
在半平面
R
e
(
s
)
>
c
Re(s)>c
Re(s)>c上一定存在，右端的积分在
R
e
(
s
)
≥
c
1
>
c
Re(s)\ge c_1>c
Re(s)≥c1>c上绝对收敛而且一致收敛。
证明
由条件
2
2
2可知，
∀
t
∈
[
0
,
+
∞
)
\forall \,t\in[0,+\infty)
∀t∈[0,+∞)，有
∣
f
(
t
)
e
−
s
t
∣
=
∣
f
(
t
)
∣
e
−
β
t
≤
M
e
−
(
β
−
c
)
t
,
R
e
(
s
)
=
β
\left|f(t)e^{-st}\right|=|f(t)|e^{-\beta t}\le Me^{-(\beta -c)t},Re(s)=\beta
∣∣f(t)e−st∣∣=∣f(t)∣e−βt≤Me−(β−c)t,Re(s)=β
若令
β
−
c
≥
ε
>
0
(
即
β
≥
c
+
ε
=
c
1
≥
c
)
\beta-c\ge\varepsilon>0(即\beta \ge c+\varepsilon=c_1\ge c)
β−c≥ε>0(即β≥c+ε=c1≥c)，则
∫
0
+
∞
∣
f
(
t
)
e
−
s
t
∣
d
t
≤
∫
0
+
∞
M
e
−
ε
t
d
t
=
M
ε
<
+
∞
\int_{0}^{+\infty}\left|{f(t)e^{-st}}\right|\,{\rm d}t \le\int_{0}^{+\infty}{Me^{-\varepsilon t}}\,{\rm d}t=\frac{M}{\varepsilon}<+\infty
∫0+∞∣∣f(t)e−st∣∣dt≤∫0+∞Me−εtdt=εM<+∞
2.1.4 常见函数的Laplace变换
L [ 1 ] = 1 s \mathscr{L}[1]=\frac{1}{s} L[1]=s1
L [ t n ] = n ! s n + 1 ( n = 1 , 2 , 3 , ⋯ ) \mathscr{L}[t^n]=\frac{n!}{s^{n+1}}(n=1,2,3,\cdots) L[tn]=sn+1n!(n=1,2,3,⋯)
L [ e a t ] = 1 s − a \mathscr{L}\left[e^{at}\right]=\frac{1}{s-a} L[eat]=s−a1
L [ sin k t ] = k s 2 + k 2 \mathscr{L}\left[\sin kt\right]=\frac{k}{s^2+k^2} L[sinkt]=s2+k2k
L [ cos k t ] = s s 2 + k 2 \mathscr{L}\left[\cos kt\right]=\frac{s}{s^2+k^2} L[coskt]=s2+k2s
L [ sinh k t ] = k s 2 − k 2 \mathscr{L}\left[\sinh kt\right]=\frac{k}{s^2-k^2} L[sinhkt]=s2−k2k
L
[
cosh
k
t
]
=
s
s
2
−
k
2
\mathscr{L}\left[\cosh kt\right]=\frac{s}{s^2-k^2}
L[coshkt]=s2−k2s
具体可查一些教科书的
L
a
p
l
a
c
e
Laplace
Laplace变换表
2.2 Laplace变换的性质
2.2.1线性性质
若
α
,
β
\alpha,\beta
α,β是常数
L
[
f
1
(
t
)
]
=
F
1
(
s
)
,
L
[
f
2
(
t
)
]
=
F
2
(
s
)
\mathscr{L}\left[f_1(t)\right]=F_1(s),\mathscr{L}\left[f_2(t)\right]=F_2(s)
L[f1(t)]=F1(s),L[f2(t)]=F2(s)
则有
L [ α f 1 ( t ) + β f 2 ( t ) ] = α L [ f 1 ( t ) ] + β L [ f 2 ( t ) ] L − 1 [ α F 1 ( s ) + β F 2 ( s ) ] = α L − 1 [ F 1 ( s ) ] + β L − 1 [ F 2 ( s ) ] } (2.2) \left.\begin{array}{l}\mathscr{L}\left[\alpha f_{1}(t)+\beta f_{2}(t)\right]&=\alpha \mathscr{L}\left[f_{1}(t)\right]+\beta \mathscr{L}\left[f_2(t)\right] \\ \mathscr{L}^{-1}\left[\alpha F_{1}(s)+\beta F_{2}(s)\right] &=\alpha \mathscr{L}^{-1}\left[F_{1}(s)\right]+\beta \mathscr{L}^{-1}\left[F_{2}(s)\right]\end{array}\right\}\tag{2.2} L[αf1(t)+βf2(t)]L−1[αF1(s)+βF2(s)]=αL[f1(t)]+βL[f2(t)]=αL−1[F1(s)]+βL−1[F2(s)]}(2.2)
2.2.2 微分性质
若
L
[
f
(
t
)
]
=
F
(
s
)
\mathscr{L}\left[f(t)\right]=F(s)
L[f(t)]=F(s)
则有
L
[
f
′
(
t
)
]
=
s
F
(
s
)
−
f
(
0
)
(2.3)
\mathscr{L}\left[f^{\prime}(t)\right]=sF(s)-f(0) \tag{2.3}
L[f′(t)]=sF(s)−f(0)(2.3)
证明
根据
L
a
p
l
a
c
e
Laplace
Laplace变换的定义，有
L
[
f
′
(
t
)
]
=
∫
0
+
∞
f
′
(
t
)
e
−
s
t
d
t
=
f
(
t
)
e
−
s
t
∣
0
+
∞
+
s
∫
0
+
∞
f
(
t
)
e
−
s
t
d
t
=
s
F
(
s
)
−
f
(
0
)
\begin{aligned}\mathscr{L}\left[f^{\prime}(t)\right]&=\int_{0}^{+\infty}{f^{\prime}(t)e^{-st}}\,{\rm d}t\\&=\left.f(t)e^{-st}\right|_{0}^{+\infty}+s\int_{0}^{+\infty}{f(t)e^{-st}}\,{\rm d}t \\&=sF(s)-f(0) \end{aligned}
L[f′(t)]=∫0+∞f′(t)e−stdt=f(t)e−st∣∣0+∞+s∫0+∞f(t)e−stdt=sF(s)−f(0)
推论
若
L
[
f
(
t
)
]
=
F
(
s
)
\mathscr{L}\left[f(t)\right]=F(s)
L[f(t)]=F(s)，则有
L
[
f
′
′
(
t
)
]
=
s
2
F
(
s
)
−
s
f
(
0
)
−
f
′
(
0
)
\mathscr{L}\left[f^{\prime \prime}(t)\right]=s^2F(s)-sf(0)-f^{\prime}(0)
L[f′′(t)]=s2F(s)−sf(0)−f′(0)
一般地
L [ f ( n ) ( t ) ] = s n F ( s ) − s n − 1 f ( 0 ) − s n − 2 f ′ ( 0 ) − ⋯ − f ( n − 1 ) ( 0 ) = s n F ( s ) − ∑ i = 0 n − 1 s n − 1 − i f ( i ) ( 0 ) (2.4) \begin{aligned}\mathscr{L}\left[f^{(n)}(t)\right]&=s^nF(s)-s^{n-1}f(0)-s^{n-2}f^{\prime}(0)-\cdots-f^{(n-1)}(0)\\&=s^nF(s)-\sum_{i=0}^{n-1}{s^{n-1-i}f^{(i)}(0)}\end{aligned}\tag{2.4} L[f(n)(t)]=snF(s)−sn−1f(0)−sn−2f′(0)−⋯−f(n−1)(0)=snF(s)−i=0∑n−1sn−1−if(i)(0)(2.4)
当
f
(
0
)
=
f
′
(
0
)
=
⋯
=
f
(
n
−
1
)
(
0
)
=
0
f(0)=f^{\prime}(0)=\cdots=f^{(n-1)}(0)=0
f(0)=f′(0)=⋯=f(n−1)(0)=0时，有
L
[
f
′
(
t
)
]
=
s
F
(
s
)
,
L
[
f
′
′
(
t
)
]
=
s
2
F
(
s
)
,
⋯
,
L
[
f
(
n
)
(
t
)
]
=
s
(
n
)
F
(
s
)
(2.5)
\mathscr{L}\left[f^{\prime}(t)\right]=sF(s), \mathscr{L}\left[f^{\prime\prime}(t)\right]=s^2F(s),\cdots, \mathscr{L}\left[f^{(n)}(t)\right]=s^{(n)}F(s)\tag{2.5}
L[f′(t)]=sF(s),L[f′′(t)]=s2F(s),⋯,L[f(n)(t)]=s(n)F(s)(2.5)
对于象函数
若
L
[
f
(
t
)
]
=
F
(
s
)
\mathscr{L}\left[f(t)\right]=F(s)
L[f(t)]=F(s)，则
F
′
(
s
)
=
−
L
[
t
f
(
t
)
]
,
R
e
(
s
)
>
c
(2.6)
F^{\prime}(s)=-\mathscr{L}\left[tf(t)\right],Re(s)>c\tag{2.6}
F′(s)=−L[tf(t)],Re(s)>c(2.6)
一般地，有
F
(
n
)
(
s
)
=
(
−
1
)
n
L
[
t
n
f
(
t
)
]
,
R
e
(
s
)
>
c
(2.7)
F^{(n)}(s)=(-1)^n\mathscr{L}\left[t^nf(t)\right],Re(s)>c\tag{2.7}
F(n)(s)=(−1)nL[tnf(t)],Re(s)>c(2.7)
2.2.3 积分性质
若
L
[
f
(
t
)
]
=
F
(
s
)
\mathscr{L}\left[f(t)\right]=F(s)
L[f(t)]=F(s)，则
L
[
∫
0
t
f
(
t
)
d
t
]
=
1
s
F
(
s
)
(2.8)
\mathscr{L}\left[\int_{0}^{t} {f(t)}\,{\rm d}t\right]=\frac{1}{s}F(s)\tag{2.8}
L[∫0tf(t)dt]=s1F(s)(2.8)
证明
设
h
(
t
)
=
∫
0
t
f
(
t
)
d
t
h(t)=\int_{0}^{t}f(t)\,{\rm d}t
h(t)=∫0tf(t)dt，则有
h
′
(
t
)
=
f
(
t
)
,
且
h
(
0
)
=
0
h^{\prime}(t)=f(t),且 h(0)=0
h′(t)=f(t),且h(0)=0
由微分性质，有
L
[
h
′
(
t
)
]
=
s
L
[
h
(
t
)
]
−
h
(
0
)
=
s
L
[
h
(
t
)
]
\mathscr{L}\left[h^{\prime}(t)\right]=s\mathscr{L}\left[h(t)\right]-h(0)= s\mathscr{L}\left[h(t)\right]
L[h′(t)]=sL[h(t)]−h(0)=sL[h(t)]
即
L
[
∫
0
t
f
(
t
)
d
t
]
=
1
s
L
[
f
(
t
)
]
=
1
s
F
(
s
)
\mathscr{L}\left[\int_{0}^{t} {f(t)}\,{\rm d}t\right] =\frac{1}{s}\mathscr{L}\left[f(t)\right]= \frac{1}{s}F(s)
L[∫0tf(t)dt]=s1L[f(t)]=s1F(s)
推广
L
{
∫
0
t
d
t
∫
0
t
d
t
⋯
∫
0
t
f
(
t
)
d
t
}
=
1
s
n
F
(
s
)
(2.9)
\mathscr{L} \left\{ \int_{0}^{t}\,{\rm d}t \int_{0}^{t}\,{\rm d}t\cdots \int_{0}^{t}f(t)\,{\rm d}t\right\}=\frac{1}{s^n}F(s)\tag{2.9}
L{∫0tdt∫0tdt⋯∫0tf(t)dt}=sn1F(s)(2.9)
象函数积分性质
L
[
f
(
t
)
t
]
=
∫
t
∞
F
(
s
)
d
s
(2.10)
\mathscr{L}\left[\frac{f(t)}{t}\right]=\int_{t}^{\infty}F(s)\,{\rm d}s\tag{2.10}
L[tf(t)]=∫t∞F(s)ds(2.10)
或
f
(
t
)
=
t
L
−
1
[
∫
t
∞
F
(
s
)
d
s
]
f(t)=t\mathscr{L}^{-1}\left[\int_{t}^{\infty}F(s)\,{\rm d}s\right]
f(t)=tL−1[∫t∞F(s)ds]
一般地，有
L
[
f
(
t
)
t
n
]
=
∫
t
∞
d
s
∫
t
∞
d
s
⋯
∫
t
∞
F
(
s
)
d
s
(2.11)
\mathscr{L}\left[\frac{f(t)}{t^n}\right]= \int_{t}^{\infty}\,{\rm d}s \int_{t}^{\infty}\,{\rm d}s\cdots \int_{t}^{\infty}F(s)\,{\rm d}s\tag{2.11}
L[tnf(t)]=∫t∞ds∫t∞ds⋯∫t∞F(s)ds(2.11)
2.2.4 位移性质
若
L
[
f
(
t
)
]
=
F
(
s
)
\mathscr{L}\left[f(t)\right]=F(s)
L[f(t)]=F(s)，则有
L
[
e
a
t
f
(
t
)
]
=
F
(
s
−
a
)
(
R
e
(
s
−
a
)
>
c
)
(2.12)
\mathscr{L}\left[e^{at}f(t)\right]=F(s-a)(Re(s-a)>c)\tag{2.12}
L[eatf(t)]=F(s−a)(Re(s−a)>c)(2.12)
证明
根据
(
2.1
)
(2.1)
(2.1)式，有
L
[
e
a
t
f
(
t
)
]
=
∫
0
+
∞
e
a
t
f
(
t
)
e
−
s
t
d
t
=
∫
0
+
∞
f
(
t
)
e
−
(
s
−
a
)
t
d
t
=
F
(
s
−
a
)
(
R
e
(
s
−
a
)
>
c
)
\mathscr{L}\left[e^{at}f(t)\right]=\int_{0}^{+\infty}e^{at}f(t)e^{-st}\,{\rm d}t=\int_{0}^{+\infty}{f(t)e^{-(s-a)t}}\,{\rm d}t=F(s-a)(Re(s-a)>c)
L[eatf(t)]=∫0+∞eatf(t)e−stdt=∫0+∞f(t)e−(s−a)tdt=F(s−a)(Re(s−a)>c)
2.2.5 延迟性质
若 L [ f ( t ) ] = F ( s ) \mathscr{L}\left[f(t)\right]=F(s) L[f(t)]=F(s)。又 t < 0 t<0 t<0时 f ( t ) = 0 f(t)=0 f(t)=0，对于任一非负实数 τ \tau τ，有
L [ f ( t − τ ) ] = e − s τ F ( s ) L − 1 [ e − s τ F ( s ) ] = f ( t − τ ) } (2.13) \left.\begin{array}{l}\mathscr{L}&\left[f(t-\tau)\right]&=e^{-s\tau}F(s) \\ \mathscr{L}^{-1}&\left[e^{-s\tau}F(s)\right]&=f(t-\tau)\end{array}\right\}\tag{2.13} LL−1[f(t−τ)][e−sτF(s)]=e−sτF(s)=f(t−τ)}(2.13)
证明
根据
(
2.1
)
(2.1)
(2.1)式有
L
[
f
(
t
−
τ
)
]
=
∫
0
+
∞
f
(
t
−
τ
)
e
−
s
t
d
t
=
∫
0
t
f
(
t
−
τ
)
e
−
s
t
d
t
+
∫
t
+
∞
f
(
t
−
τ
)
e
−
s
t
d
t
\mathscr{L}\left[f(t-\tau)\right]=\int_{0}^{+\infty}{f(t-\tau)e^{-st}}\,{\rm d}t=\int_{0}^{t}{f(t-\tau)e^{-st}}\,{\rm d}t+ \int_{t}^{+\infty}{f(t-\tau)e^{-st}}\,{\rm d}t
L[f(t−τ)]=∫0+∞f(t−τ)e−stdt=∫0tf(t−τ)e−stdt+∫t+∞f(t−τ)e−stdt
令
t
−
τ
=
u
t-\tau=u
t−τ=u，则
L
[
f
(
t
−
τ
)
]
=
∫
0
+
∞
f
(
u
)
e
−
s
(
u
+
τ
)
d
u
=
e
−
s
τ
∫
0
+
∞
f
(
u
)
e
−
s
u
d
u
=
e
−
s
τ
F
(
s
)
(
R
e
(
s
)
>
c
)
\begin{aligned} \mathscr{L}\left[f(t-\tau)\right]&=\int_{0}^{+\infty}{f(u)e^{-s(u+\tau)}}\,{\rm d}u\\&=e^{-s\tau}\int_{0}^{+\infty}{f(u)e^{-su}}\,{\rm d}u\\&=e^{-s\tau}F(s) (Re(s)>c) \end{aligned}
L[f(t−τ)]=∫0+∞f(u)e−s(u+τ)du=e−sτ∫0+∞f(u)e−sudu=e−sτF(s)(Re(s)>c)
2.2.6 初值定理与终值定理
初值定理
若
L
[
f
(
t
)
]
=
F
(
s
)
\mathscr{L}\left[f(t)\right]=F(s)
L[f(t)]=F(s)，且
lim
s
→
∞
s
F
(
s
)
\lim\limits_{s \to \infty}{sF(s)}
s→∞limsF(s)存在，则
lim
t
→
0
f
(
t
)
=
lim
s
→
∞
s
F
(
s
)
o
r
f
(
0
)
=
lim
s
→
∞
s
F
(
s
)
}
(2.14)
\left.\begin{array}{l}\lim\limits_{t \to 0}&f(t)&=\lim\limits_{s \to \infty}sF(s) \\or\\&f(0)&=\lim\limits_{s\to \infty}sF(s)\end{array}\right\}\tag{2.14}
t→0limorf(t)f(0)=s→∞limsF(s)=s→∞limsF(s)⎭⎪⎬⎪⎫(2.14)
证明
根据
L
a
p
l
a
c
e
Laplace
Laplace变换的微分性质
L
[
f
′
(
t
)
]
=
s
L
[
f
(
t
)
]
−
f
(
0
)
=
s
F
(
s
)
−
f
(
0
)
\begin{aligned} \mathscr{L}\left[f^{\prime}(t)\right]&=s\mathscr{L}\left[f(t)\right]-f(0)\\&=sF(s)-f(0) \end{aligned}
L[f′(t)]=sL[f(t)]−f(0)=sF(s)−f(0)
若
lim
s
→
∞
s
F
(
s
)
\lim\limits_{s \to \infty}{sF(s)}
s→∞limsF(s)存在，则
lim
R
e
(
s
)
→
+
∞
s
F
(
s
)
\lim\limits_{Re(s) \to +\infty}{sF(s)}
Re(s)→+∞limsF(s)必存在，且两者相等。
lim
s
→
∞
s
F
(
s
)
=
lim
R
e
(
s
)
→
+
∞
s
F
(
s
)
\lim\limits_{s \to \infty}{sF(s)}= \lim\limits_{Re(s) \to +\infty}{sF(s)}
s→∞limsF(s)=Re(s)→+∞limsF(s)
两边取
R
e
(
s
)
→
+
∞
Re(s)\to +\infty
Re(s)→+∞时的极限
lim
R
e
(
s
)
→
+
∞
L
[
f
′
(
t
)
]
=
lim
R
e
(
s
)
→
+
∞
[
s
L
[
f
(
t
)
]
−
f
(
0
)
]
=
lim
s
→
∞
s
F
(
s
)
−
f
(
0
)
\begin{aligned} \lim\limits_{Re(s)\to+\infty}\mathscr{L}\left[f^{\prime}(t)\right]&= \lim\limits_{Re(s)\to+\infty} \left[s\mathscr{L}\left[f(t)\right]-f(0)\right]\\&= \lim\limits_{s\to\infty} sF(s)-f(0) \end{aligned}
Re(s)→+∞limL[f′(t)]=Re(s)→+∞lim[sL[f(t)]−f(0)]=s→∞limsF(s)−f(0)
另一方面
lim
R
e
(
s
)
→
+
∞
L
[
f
′
(
t
)
]
=
lim
R
e
(
s
)
→
+
∞
∫
0
+
∞
f
′
(
t
)
e
−
s
t
d
t
=
∫
0
+
∞
lim
R
e
(
s
)
→
+
∞
f
′
(
t
)
e
−
s
t
d
t
=
0
\begin{aligned} \lim\limits_{Re(s)\to+\infty}\mathscr{L}\left[f^{\prime}(t)\right]&= \lim\limits_{Re(s)\to+\infty} \int_{0}^{+\infty}{f^{\prime}(t)e^{-st}}\,{\rm d}t\\&=\int_{0}^{+\infty} \lim\limits_{Re(s)\to+\infty} f^{\prime}(t)e^{-st}\,{\rm d}t\\&=0 \end{aligned}
Re(s)→+∞limL[f′(t)]=Re(s)→+∞lim∫0+∞f′(t)e−stdt=∫0+∞Re(s)→+∞limf′(t)e−stdt=0
所以
lim
s
→
∞
s
F
(
s
)
−
f
(
0
)
=
0
\lim\limits_{s\to\infty} sF(s)-f(0)=0
s→∞limsF(s)−f(0)=0
lim
s
→
∞
s
F
(
s
)
=
f
(
0
)
=
lim
t
→
0
f
(
t
)
\lim\limits_{s\to\infty} sF(s)=f(0)=\lim\limits_{t\to 0}f(t)
s→∞limsF(s)=f(0)=t→0limf(t)
终值定理
若
L
[
f
(
t
)
]
=
F
(
s
)
\mathscr{L}\left[f(t)\right]=F(s)
L[f(t)]=F(s)，且
s
F
(
s
)
sF(s)
sF(s)的所有奇点全在
s
s
s平面的左半部，则
lim
t
→
+
∞
f
(
t
)
=
lim
s
→
0
s
F
(
s
)
o
r
f
(
+
∞
)
=
lim
s
→
0
s
F
(
s
)
}
(2.15)
\left.\begin{array}{l}&\lim\limits_{t\to+\infty}f(t)&=\lim\limits_{s\to0}sF(s) \\or\\&f(+\infty)&= \lim\limits_{s\to0}sF(s)\end{array}\right\}\tag{2.15}
ort→+∞limf(t)f(+∞)=s→0limsF(s)=s→0limsF(s)⎭⎪⎬⎪⎫(2.15)
证明
根据定理给出条件与微分性质
L
[
f
′
(
t
)
]
=
s
F
(
s
)
−
f
(
0
)
\mathscr{L}\left[f^{\prime}(t)\right]=sF(s)-f(0)
L[f′(t)]=sF(s)−f(0)
两边取
s
→
0
s\to 0
s→0的极限，得
lim
s
→
0
L
[
f
′
(
t
)
]
=
lim
s
→
0
[
s
F
(
s
)
−
f
(
0
)
]
=
lim
s
→
0
s
F
(
s
)
−
f
(
0
)
\lim\limits_{s\to 0}\mathscr{L} \left[f^{\prime}(t)\right]=\lim\limits_{s\to 0}\left[sF(s)-f(0)\right]= \lim\limits_{s\to 0} sF(s)-f(0)
s→0limL[f′(t)]=s→0lim[sF(s)−f(0)]=s→0limsF(s)−f(0)
而
lim
s
→
0
L
[
f
′
(
t
)
]
=
lim
s
→
0
∫
0
+
∞
f
′
(
t
)
e
−
s
t
d
s
=
∫
0
+
∞
lim
s
→
0
e
−
s
t
f
′
(
t
)
d
t
=
∫
0
+
∞
f
′
(
t
)
d
t
=
lim
t
→
+
∞
f
(
t
)
−
f
(
0
)
\begin{aligned} \lim\limits_{s\to 0}\mathscr{L}\left[f^{\prime}(t)\right]&=\lim\limits_{s \to 0}\int_{0}^{+\infty}{f^{\prime}(t)e^{-st}}\,{\rm d}s\\&=\int_{0}^{+\infty}{\lim\limits_{s\to 0}e^{-st}f^{\prime}(t)}\,{\rm d}t\\&=\int_{0}^{+\infty}{f^{\prime}(t)}\,{\rm d}t\\&= \lim\limits_{t\to +\infty} f(t)-f(0) \end{aligned}
s→0limL[f′(t)]=s→0lim∫0+∞f′(t)e−stds=∫0+∞s→0lime−stf′(t)dt=∫0+∞f′(t)dt=t→+∞limf(t)−f(0)
所以
lim
t
→
+
∞
f
(
t
)
−
f
(
0
)
=
lim
s
→
0
s
F
(
s
)
−
f
(
0
)
\lim\limits_{t\to +\infty} f(t)-f(0) =\lim\limits_{s\to 0} sF(s)-f(0)
t→+∞limf(t)−f(0)=s→0limsF(s)−f(0)
即
lim
t
→
+
∞
f
(
t
)
=
f
(
+
∞
)
=
lim
s
→
0
s
F
(
s
)
\lim\limits_{t\to +\infty} f(t)=f(+\infty)= \lim\limits_{s\to 0} sF(s)
t→+∞limf(t)=f(+∞)=s→0limsF(s)
2.3 Laplace逆变换
函数
f
(
t
)
f(t)
f(t)的
L
a
p
l
a
c
e
Laplace
Laplace变换，实际上是
f
(
t
)
u
(
t
)
e
−
β
t
f(t)u(t)e^{-\beta t}
f(t)u(t)e−βt的
F
o
u
r
i
e
r
Fourier
Fourier变换。于是，当
f
(
t
)
u
(
t
)
e
−
β
t
f(t)u(t)e^{-\beta t}
f(t)u(t)e−βt满足
F
o
u
r
i
e
r
Fourier
Fourier积分定理的条件时，按
F
o
u
r
i
e
r
Fourier
Fourier积分公式，在
f
(
t
)
f(t)
f(t)连续点处有
f
(
t
)
u
(
t
)
e
−
β
t
=
1
2
π
∫
−
∞
+
∞
[
∫
−
∞
+
∞
f
(
τ
)
u
(
τ
)
e
−
β
τ
e
−
i
ω
t
d
τ
]
e
i
ω
t
d
ω
=
1
2
π
∫
−
∞
+
∞
e
i
ω
t
d
ω
[
∫
0
+
∞
f
(
τ
)
e
−
(
β
+
i
ω
)
τ
d
t
]
=
1
2
π
∫
−
∞
+
∞
F
(
β
+
i
ω
)
e
i
ω
t
d
ω
,
t
>
0
\begin{aligned} f(t)u(t)e^{-\beta t}&=\frac{1}{2\pi}\int_{-\infty}^{+\infty}{\left[\int_{-\infty}^{+\infty} f(\tau)u(\tau)e^{-\beta \tau}e^{- i\omega t}\,{\rm d}\tau\right]e^{i\omega t}}\,{\rm d}\omega\\&=\frac{1}{2\pi}\int_{-\infty}^{+\infty}{e^{i\omega t}}\,{\rm d}\omega\left[\int_{0}^{+\infty}{f(\tau)e^{-(\beta+i\omega)\tau}}\,{\rm d}t\right]\\&=\frac{1}{2\pi}\int_{-\infty}^{+\infty}{F(\beta+i\omega)e^{i\omega t}}\,{\rm d}\omega,t>0 \end{aligned}
f(t)u(t)e−βt=2π1∫−∞+∞[∫−∞+∞f(τ)u(τ)e−βτe−iωtdτ]eiωtdω=2π1∫−∞+∞eiωtdω[∫0+∞f(τ)e−(β+iω)τdt]=2π1∫−∞+∞F(β+iω)eiωtdω,t>0
等式两边乘以
e
β
t
e^{\beta t}
eβt，令
β
+
i
ω
=
s
\beta+i\omega=s
β+iω=s，有
f
(
t
)
=
1
2
π
i
∫
β
−
i
∞
β
+
i
∞
F
(
s
)
e
s
t
d
s
(
t
>
0
)
(2.16)
f(t)=\frac{1}{2\pi i}\int_{\beta-i\infty}^{\beta+i\infty}{F(s)e^{st}}\,{\rm d}s(t>0)\tag{2.16}
f(t)=2πi1∫β−i∞β+i∞F(s)estds(t>0)(2.16)
此为象函数
F
(
s
)
F(s)
F(s)求它的象原函数
f
(
t
)
f(t)
f(t)的一般公式，右端积分称为
L
a
p
l
a
c
e
Laplace
Laplace反演积分。
借助留数定理计算
若
s
1
,
s
2
,
⋯
,
s
n
s_1,s_2,\cdots,s_n
s1,s2,⋯,sn是函数
F
(
s
)
F(s)
F(s)的所有奇点(适当选取
β
\beta
β使这些奇点全在
R
e
(
s
)
<
β
Re(s)<\beta
Re(s)<β的范围内)，且当
s
→
∞
s \to \infty
s→∞时，
F
(
s
)
→
0
F(s)\to0
F(s)→0，则有
f
(
t
)
=
1
2
π
i
∫
β
−
i
∞
β
+
i
∞
F
(
s
)
e
s
t
d
s
=
∑
k
=
1
n
Res
s
=
s
k
[
F
(
s
)
e
s
t
]
,
(
t
>
0
)
(2.17)
f(t)= \frac{1}{2\pi i}\int_{\beta-i\infty}^{\beta+i\infty}{F(s)e^{st}}\,{\rm d}s=\sum_{k=1}^{n}{\underset{s=s_{k}}{\text{Res}}\left[F(s)e^{st}\right]},(t>0)\tag{2.17}
f(t)=2πi1∫β−i∞β+i∞F(s)estds=k=1∑ns=skRes[F(s)est],(t>0)(2.17)
证明用到了
J
o
r
d
a
n
Jordan
Jordan定理，可查阅复变函数论
除了上述方法，还可以用部分分式和查表方法解决。
2.4 卷积
假定
f
1
(
t
)
,
f
2
(
t
)
f_1(t),f_2(t)
f1(t),f2(t)满足
L
a
p
l
a
c
e
Laplace
Laplace变换存在定理中的条件，且
L
[
f
1
(
t
)
]
=
F
1
(
s
)
,
L
[
f
2
(
t
)
]
=
F
2
(
s
)
\mathscr{L}\left[f_1(t)\right]=F_1(s), \mathscr{L}\left[f_2(t)\right]=F_2(s)
L[f1(t)]=F1(s),L[f2(t)]=F2(s)，则
f
1
(
t
)
∗
f
2
(
t
)
f_1(t)\ast f_2(t)
f1(t)∗f2(t)的
L
a
p
l
a
c
e
Laplace
Laplace变换一定存在，且
L
[
f
1
(
t
)
∗
f
2
(
t
)
]
=
F
1
(
s
)
⋅
F
2
(
s
)
o
r
L
−
1
[
F
1
(
s
)
⋅
F
2
(
s
)
]
=
f
1
(
t
)
∗
f
2
(
t
)
}
(2.18)
\left.\begin{array}{l}\mathscr{L}\left[f_1(t)\ast f_2(t)\right]&=F_1(s)\cdot F_2(s)\\or \\\mathscr{L}^{-1}\left[F_1(s)\cdot F_2(s)\right]&=f_1(t)\ast f_2(t)\end{array}\right\}\tag{2.18}
L[f1(t)∗f2(t)]orL−1[F1(s)⋅F2(s)]=F1(s)⋅F2(s)=f1(t)∗f2(t)⎭⎬⎫(2.18)
推广
若
f
k
(
t
)
(
k
=
1
,
2
,
⋯
,
n
)
f_k(t)(k=1,2,\cdots,n)
fk(t)(k=1,2,⋯,n)满足
L
a
p
l
a
c
e
Laplace
Laplace变换存在条件，且
L
[
f
k
(
t
)
]
=
F
k
(
s
)
(
k
=
1
,
2
,
⋯
,
n
)
\mathscr{L}\left[f_k(t)\right]=F_k(s)(k=1,2,\cdots,n)
L[fk(t)]=Fk(s)(k=1,2,⋯,n)，则有
L
[
f
1
(
t
)
∗
f
2
(
t
)
∗
⋯
∗
f
n
(
t
)
]
=
F
1
(
s
)
⋅
F
2
(
s
)
⋅
⋯
⋅
F
n
(
s
)
(2.19)
\mathscr{L}\left[f_1(t)\ast f_2(t)\ast \cdots\ast f_n(t)\right]=F_1(s)\cdot F_2(s)\cdot \cdots \cdot F_n(s)\tag{2.19}
L[f1(t)∗f2(t)∗⋯∗fn(t)]=F1(s)⋅F2(s)⋅⋯⋅Fn(s)(2.19)
扫一扫
1万+
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
