线性系统粗浅认识——第三次作业
作者:互联网
线性系统粗浅认识——第三次作业
声明:本人特别菜,不研究相关的方向,差点挂科,这个作业的内容仅供交流。
第三次作业
作业1
题目
若 z z z是 X X X的子空间, A A A为方阵,对任意 x ∈ z x \in z x∈z, A x ∈ z Ax \in z Ax∈z ,则称 是 的A不变子空间,补齐例子中的所有内容
作业1解答_例子
X
X
X三维空间,
z
z
z是
X
X
X的三维子空间,基底
(
1
1
0
)
\begin{pmatrix} 1\\1\\0\end{pmatrix}
⎝⎛110⎠⎞
(
0
1
0
)
\begin{pmatrix} 0\\1\\0\end{pmatrix}
⎝⎛010⎠⎞,
z
z
z的任意一个元素
x
=
α
(
1
1
0
)
+
β
(
0
1
0
)
x = \alpha \left( {\begin{matrix} 1\\ 1\\ 0 \end{matrix}} \right) + \beta \left({\begin{matrix} 0\\ 1\\ 0 \end{matrix}} \right)
x=α⎝⎛110⎠⎞+β⎝⎛010⎠⎞
A
=
[
2
3
0
−
1
1
0
0
0
1
]
A = \left[ {\begin{matrix}{} 2&3&0\\ { - 1}&1&0\\ 0&0&1 \end{matrix}} \right]
A=⎣⎡2−10310001⎦⎤
A x = ( 2 3 0 − 1 1 0 0 0 1 ) × α ( 1 1 0 ) + ( 2 3 0 − 1 1 0 0 0 1 ) × β ( 0 1 0 ) = α ( 5 0 0 ) + β ( 3 1 0 ) = ( 5 α + 3 β ) ( 1 1 0 ) + ( − 5 α − 2 β ) ( 0 1 0 ) \begin{matrix}{} Ax = \left( {\begin{matrix}{} 2&3&0\\ { - 1}&1&0\\ 0&0&1 \end{matrix}} \right) \times \alpha \left( {\begin{matrix}{} 1\\ 1\\ 0 \end{matrix}} \right) + \left( {\begin{matrix}{} 2&3&0\\ { - 1}&1&0\\ 0&0&1 \end{matrix}} \right) \times \beta \left( {\begin{matrix}{} 0\\ 1\\ 0 \end{matrix}} \right)\\ = \alpha \left( {\begin{matrix}{} 5\\ 0\\ 0 \end{matrix}} \right) + \beta \left( {\begin{matrix}{} 3\\ 1\\ 0 \end{matrix}} \right)\\ = (5\alpha + 3\beta )\left( {\begin{matrix}{} 1\\ 1\\ 0 \end{matrix}} \right) + ( - 5\alpha - 2\beta )\left( {\begin{matrix}{} 0\\ 1\\ 0 \end{matrix}} \right) \end{matrix} Ax=⎝⎛2−10310001⎠⎞×α⎝⎛110⎠⎞+⎝⎛2−10310001⎠⎞×β⎝⎛010⎠⎞=α⎝⎛500⎠⎞+β⎝⎛310⎠⎞=(5α+3β)⎝⎛110⎠⎞+(−5α−2β)⎝⎛010⎠⎞
作业2
题目:证明必要性
{ A , B } \left\{ {A,B} \right\} {A,B}完全能控 完全等价于的所有列不属于任意一个 的真线性不变子空间,给出简单例子并加以说明。
分析
证明该命题的必要性比较复杂,由于原命题和原命题的逆否命题是等价的,因此我改证原命题的逆否命题。
证明
逆否命题:如果
{
A
,
B
}
\left\{ {A,B} \right\}
{A,B}不完全能控,就有
B
B
B的所有列属于一个存在的
A
A
A的真线性不变子空间。
{
A
,
B
}
\left\{ {A,B} \right\}
{A,B}不完全能控,可以得到
[
B
,
A
B
,
⋯
,
A
n
−
1
B
]
[B,AB, \cdots ,{A^{n - 1}}B]
[B,AB,⋯,An−1B]非行满秩
存在非零向量
w
w
w,使
w
[
B
,
A
B
,
⋯
,
A
n
−
1
B
]
=
0
w[B,AB, \cdots ,{A^{n - 1}}B] = 0
w[B,AB,⋯,An−1B]=0
所以
w
B
=
0
wB = 0
wB=0,所以
r
a
n
k
[
B
]
<
n
rank[B] < n
rank[B]<n
A
A
A 的特征根的
λ
1
,
λ
2
,
⋯
λ
n
{\lambda _{\rm{1}}},{\lambda _2}, \cdots {\lambda _n}
λ1,λ2,⋯λn,
A
A
A的特征向量是
v
1
,
v
2
,
⋯
v
n
{v_{\rm{1}}},{v_2}, \cdots {v_n}
v1,v2,⋯vn,多输入系统
B
B
B的向量,
B
=
[
b
1
⋯
b
m
]
b
1
=
α
11
v
1
+
⋯
+
α
1
k
v
k
⋮
b
m
=
α
m
1
v
1
+
⋯
+
α
m
k
v
k
B = \left[ {\begin{matrix}{} {{b_1}}& \cdots &{{b_m}} \end{matrix}} \right]\\ {b_1} = {\alpha _{11}}{v_1} + \cdots + {\alpha _{1k}}{v_k}\\ \vdots \\ {b_m} = {\alpha _{m1}}{v_1} + \cdots + {\alpha _{mk}}{v_k}
B=[b1⋯bm]b1=α11v1+⋯+α1kvk⋮bm=αm1v1+⋯+αmkvk
其中
k
<
n
k < n
k<n,
v
1
,
v
2
,
⋯
v
k
{v_{\rm{1}}},{v_2}, \cdots {v_k}
v1,v2,⋯vk张成了一个线性空间
A
′
A'
A′,
A
′
A'
A′ 是
A
A
A的真线性不变子空间
所以
B
B
B 的所有列属于一个存在的
A
A
A的真线性不变子空间。
所以逆否命题为真,因此,该命题为真。
直接证明:
设矩阵
A
A
A 的一组特征向量为
v
1
,
v
2
,
⋯
v
n
{v_{\rm{1}}},{v_2}, \cdots {v_n}
v1,v2,⋯vn,
B
B
B 的所有列不属于任意一个
A
A
A 的特征向量张成的真线性不变子空间,即
B
B
B的基底可以表示为
{
v
1
,
v
2
,
⋯
v
n
,
v
n
+
1
,
⋯
v
k
}
\{ {v_{\rm{1}}},{v_2}, \cdots {v_n},{v_{n + 1}}, \cdots {v_k}\}
{v1,v2,⋯vn,vn+1,⋯vk}其中
k
>
n
k > n
k>n设
B
B
B的输入为
m
m
m维也就是说
B
B
B可以表示为
B
=
[
b
1
⋯
b
m
]
b
1
=
α
11
v
1
+
⋯
+
α
1
k
v
k
⋮
b
m
=
α
m
1
v
1
+
⋯
+
α
m
k
v
k
B = \left[ {\begin{matrix}{} {{b_1}}& \cdots &{{b_m}} \end{matrix}} \right]\\ {b_1} = {\alpha _{11}}{v_1} + \cdots + {\alpha _{1k}}{v_k}\\ \vdots \\ {b_m} = {\alpha _{m1}}{v_1} + \cdots + {\alpha _{mk}}{v_k}
B=[b1⋯bm]b1=α11v1+⋯+α1kvk⋮bm=αm1v1+⋯+αmkvk
计算零状态响应
x
(
t
)
=
∫
t
0
t
e
A
(
t
−
τ
)
B
u
(
τ
)
d
τ
=
∫
t
0
t
e
A
(
t
−
τ
)
[
b
1
⋯
b
m
]
u
(
τ
)
d
τ
=
∫
t
0
t
e
A
(
t
−
τ
)
b
1
u
(
τ
)
d
τ
+
⋯
+
∫
t
0
t
e
A
(
t
−
τ
)
b
m
u
(
τ
)
d
τ
=
∫
t
0
t
e
A
(
t
−
τ
)
(
α
11
v
1
+
α
12
v
2
+
⋯
+
α
1
n
v
n
+
α
1
n
+
1
v
n
+
1
+
⋯
+
α
1
k
v
k
)
u
(
τ
)
d
τ
+
⋯
+
∫
t
0
t
e
A
(
t
−
τ
)
(
α
m
1
v
1
+
α
m
2
v
2
+
⋯
+
α
m
n
v
n
+
α
m
n
+
1
v
n
+
1
+
⋯
+
α
m
k
v
k
)
u
(
τ
)
d
τ
=
∫
t
0
t
e
A
(
t
−
τ
)
α
11
v
1
u
(
τ
)
d
τ
+
⋯
+
∫
t
0
t
e
A
(
t
−
τ
)
α
1
n
v
n
u
(
τ
)
d
τ
+
∫
t
0
t
e
A
(
t
−
τ
)
α
1
n
+
1
v
n
+
1
u
(
τ
)
d
τ
+
⋯
+
∫
t
0
t
e
A
(
t
−
τ
)
α
1
k
v
k
u
(
τ
)
d
τ
+
⋯
+
∫
t
0
t
e
A
(
t
−
τ
)
α
m
1
v
1
u
(
τ
)
d
τ
+
⋯
+
∫
t
0
t
e
A
(
t
−
τ
)
α
m
n
v
n
u
(
τ
)
d
τ
+
∫
t
0
t
e
A
(
t
−
τ
)
α
m
n
+
1
v
n
+
1
u
(
τ
)
d
τ
+
⋯
+
∫
t
0
t
e
A
(
t
−
τ
)
α
m
k
v
k
u
(
τ
)
d
τ
=
(
α
11
+
⋯
α
m
1
)
∫
t
0
t
e
λ
1
(
t
−
τ
)
v
1
u
(
τ
)
d
τ
+
⋯
+
(
α
1
n
+
⋯
+
α
m
n
)
∫
t
0
t
e
λ
n
(
t
−
τ
)
v
n
u
(
τ
)
d
τ
+
(
α
1
n
+
1
+
⋯
+
α
m
n
+
1
)
∫
t
0
t
e
A
(
t
−
τ
)
v
n
+
1
u
(
τ
)
d
τ
+
⋯
+
(
α
1
k
+
⋯
+
α
m
k
)
∫
t
0
t
e
A
(
t
−
τ
)
v
k
u
(
τ
)
d
τ
x(t) = \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}Bu(\tau )d\tau } = \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}\left[ {\begin{matrix}{} {{b_1}}& \cdots &{{b_m}} \end{matrix}} \right]u(\tau )d\tau } \\ = \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{b_1}u(\tau )d\tau } + \cdots + \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{b_m}u(\tau )d\tau } \\ = \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}({\alpha _{11}}{v_1} + {\alpha _{12}}{v_2} + \cdots + {\alpha _{1n}}{v_n} + {\alpha _{1n{\rm{ + 1}}}}{v_{n{\rm{ + 1}}}} + \cdots + {\alpha _{1k}}{v_k})u(\tau )d\tau + \cdots + } \\ \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}({\alpha _{m1}}{v_1} + {\alpha _{m2}}{v_2} + \cdots + {\alpha _{mn}}{v_n} + {\alpha _{mn{\rm{ + 1}}}}{v_{n{\rm{ + 1}}}} + \cdots + {\alpha _{mk}}{v_k})u(\tau )d\tau } \\ = \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{11}}{v_1}u(\tau )d\tau } + \cdots + \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{1n}}{v_n}u(\tau )d\tau } + \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{1n + 1}}{v_{n + 1}}u(\tau )d\tau } + \cdots + \\ \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{1k}}{v_k}u(\tau )d\tau } + \cdots + \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{m1}}{v_1}u(\tau )d\tau } + \cdots + \\ \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{mn}}{v_n}u(\tau )d\tau } + \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{mn + 1}}{v_{n + 1}}u(\tau )d\tau } + \cdots + \int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{\alpha _{mk}}{v_k}u(\tau )d\tau } \\ = ({\alpha _{11}} + \cdots {\alpha _{m1}})\int\limits_{{t_0}}^t {{e^{{\lambda _1}(t - \tau )}}{v_1}u(\tau )d\tau } + \cdots + ({\alpha _{1n}} + \cdots + {\alpha _{mn}})\int\limits_{{t_0}}^t {{e^{{\lambda _n}(t - \tau )}}{v_n}u(\tau )d\tau } + \\ ({\alpha _{1n + 1}} + \cdots + {\alpha _{mn + 1}})\int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{v_{n + 1}}u(\tau )d\tau } + \cdots + ({\alpha _{1k}} + \cdots + {\alpha _{mk}})\int\limits_{{t_0}}^t {{e^{A(t - \tau )}}{v_k}u(\tau )d\tau }
x(t)=t0∫teA(t−τ)Bu(τ)dτ=t0∫teA(t−τ)[b1⋯bm]u(τ)dτ=t0∫teA(t−τ)b1u(τ)dτ+⋯+t0∫teA(t−τ)bmu(τ)dτ=t0∫teA(t−τ)(α11v1+α12v2+⋯+α1nvn+α1n+1vn+1+⋯+α1kvk)u(τ)dτ+⋯+t0∫teA(t−τ)(αm1v1+αm2v2+⋯+αmnvn+αmn+1vn+1+⋯+αmkvk)u(τ)dτ=t0∫teA(t−τ)α11v1u(τ)dτ+⋯+t0∫teA(t−τ)α1nvnu(τ)dτ+t0∫teA(t−τ)α1n+1vn+1u(τ)dτ+⋯+t0∫teA(t−τ)α1kvku(τ)dτ+⋯+t0∫teA(t−τ)αm1v1u(τ)dτ+⋯+t0∫teA(t−τ)αmnvnu(τ)dτ+t0∫teA(t−τ)αmn+1vn+1u(τ)dτ+⋯+t0∫teA(t−τ)αmkvku(τ)dτ=(α11+⋯αm1)t0∫teλ1(t−τ)v1u(τ)dτ+⋯+(α1n+⋯+αmn)t0∫teλn(t−τ)vnu(τ)dτ+(α1n+1+⋯+αmn+1)t0∫teA(t−τ)vn+1u(τ)dτ+⋯+(α1k+⋯+αmk)t0∫teA(t−τ)vku(τ)dτ
该系统的响应可以在 v 1 v_1 v1 v 2 v_2 v2 . . . ... ... v n v_n vn方向上,所以系统 { A , B } \left\{ {A,B} \right\} {A,B}所有状态变量都可以由输入信号控制。所以该系统是完全能控的。
例子:
x
˙
=
(
2
1
0
3
)
x
+
(
0
1
)
u
\dot x = \left( {\begin{matrix}{} 2&1\\ 0&3 \end{matrix}} \right)x + \left( {\begin{matrix}{} 0\\ 1 \end{matrix}} \right)u
x˙=(2013)x+(01)u
已知该系统完全能控, A A A的特征值为 λ 1 = 2 , λ 2 = 3 {\lambda _1} = 2,{\lambda _2} = 3 λ1=2,λ2=3特征向量 v 1 = ( 1 0 ) , v 2 = ( 0 . 707 0 . 707 ) {v_1} = \left( {\begin{matrix}{} 1\\ 0 \end{matrix}} \right),{v_2} = \left( {\begin{matrix}{} {{\rm{0}}{\rm{.707}}}\\ {{\rm{0}}{\rm{.707}}} \end{matrix}} \right) v1=(10),v2=(0.7070.707)
B = − v 1 + 1 0.707 v 2 = − ( 1 0 ) + 1 0.707 × ( 0.707 0.707 ) = ( 0 1 ) B = - {v_1} + \frac{1}{{0.707}}{v_2} = {\rm{ - }}\left( {\begin{matrix}{} {\rm{1}}\\ {\rm{0}} \end{matrix}} \right){\rm{ + }}\frac{1}{{0.707}} \times \left( {\begin{matrix}{} {0.707}\\ {0.707} \end{matrix}} \right){\rm{ = }}\left( {\begin{matrix}{} {\rm{0}}\\ {\rm{1}} \end{matrix}} \right) B=−v1+0.7071v2=−(10)+0.7071×(0.7070.707)=(01)
其中 B B B的所有列不属于任意一个 A A A的真线性不变子空间, B B B的所有列属于 的线性不变子空间。
作业3
题目
以4阶两输入系统为例,分析 A A A的特征向量和 B B B的每列之间的关系,说明系统的能控性。
解答
A
A
A的特征根的
λ
1
,
λ
2
,
λ
3
,
λ
4
{\lambda _{\rm{1}}},{\lambda _2},{\lambda _3},{\lambda _4}
λ1,λ2,λ3,λ4,
A
A
A 的特征向量是
v
1
,
v
2
,
v
3
,
v
4
{v_{\rm{1}}},{v_2},{v_3},{v_4}
v1,v2,v3,v4 ,两输入系统 的向量
B
=
[
b
1
b
2
]
B = \left[ {\begin{matrix}{} {{b_1}}&{{b_2}} \end{matrix}} \right]
B=[b1b2],
b
1
=
α
11
v
1
+
⋯
+
α
1
k
v
k
b
2
=
α
21
v
1
+
⋯
+
α
2
k
v
k
\begin{array}{l} {b_1} = {\alpha _{11}}{v_1} + \cdots + {\alpha _{1k}}{v_k}\\ {b_2} = {\alpha _{21}}{v_1} + \cdots + {\alpha _{2k}}{v_k} \end{array}
b1=α11v1+⋯+α1kvkb2=α21v1+⋯+α2kvk
x
(
t
)
=
e
A
t
x
(
0
)
+
∫
0
t
e
A
(
t
−
τ
)
B
u
(
τ
)
d
τ
x(t) = {e^{At}}x(0) + \int\limits_0^t {{e^{A(t - \tau )}}} Bu(\tau )d\tau
x(t)=eAtx(0)+0∫teA(t−τ)Bu(τ)dτ
∫
0
t
e
A
(
t
−
τ
)
B
u
(
τ
)
d
τ
=
∫
0
t
e
A
(
t
−
τ
)
[
b
1
b
2
]
u
(
τ
)
d
τ
=
∫
0
t
e
A
(
t
−
τ
)
(
α
11
v
1
+
⋯
+
α
1
k
v
k
)
u
(
τ
)
d
τ
+
∫
0
t
e
A
(
t
−
τ
)
(
α
21
v
1
+
⋯
+
α
2
k
v
k
)
u
(
τ
)
d
τ
=
(
α
11
+
α
21
)
∫
0
t
e
λ
1
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
1
+
⋯
+
(
α
1
k
+
α
2
k
)
∫
0
t
e
A
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
k
\int\limits_0^t {{e^{A(t - \tau )}}} Bu(\tau )d\tau = \int\limits_0^t {{e^{A(t - \tau )}}} \left[ {\begin{matrix}{} {{b_1}}&{{b_2}} \end{matrix}} \right]u(\tau )d\tau \\ = \int\limits_0^t {{e^{A(t - \tau )}}} ({\alpha _{11}}{v_1} + \cdots + {\alpha _{1k}}{v_k})u(\tau )d\tau + \int\limits_0^t {{e^{A(t - \tau )}}} ({\alpha _{21}}{v_1} + \cdots + {\alpha _{2k}}{v_k})u(\tau )d\tau \\ = ({\alpha _{11}} + {\alpha _{21}})\int\limits_0^t {{e^{{\lambda _1}(t - \tau )}}} u(\tau )d\tau \bullet {v_1} + \cdots + ({\alpha _{1k}} + {\alpha _{2k}})\int\limits_0^t {{e^{A(t - \tau )}}} u(\tau )d\tau \bullet {v_k}
0∫teA(t−τ)Bu(τ)dτ=0∫teA(t−τ)[b1b2]u(τ)dτ=0∫teA(t−τ)(α11v1+⋯+α1kvk)u(τ)dτ+0∫teA(t−τ)(α21v1+⋯+α2kvk)u(τ)dτ=(α11+α21)0∫teλ1(t−τ)u(τ)dτ∙v1+⋯+(α1k+α2k)0∫teA(t−τ)u(τ)dτ∙vk
当 的时候
k
<
4
k < 4
k<4,上式可以表示为
∫
0
t
e
A
(
t
−
τ
)
B
u
(
τ
)
d
τ
=
(
α
11
+
α
21
)
∫
0
t
e
λ
1
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
1
+
⋯
+
(
α
1
k
+
α
2
k
)
∫
0
t
e
A
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
4
=
(
α
11
+
α
21
)
∫
0
t
e
λ
1
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
1
+
⋯
+
(
α
1
k
+
α
2
k
)
∫
0
t
e
λ
k
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
k
\int\limits_0^t {{e^{A(t - \tau )}}} Bu(\tau )d\tau \\ = ({\alpha _{11}} + {\alpha _{21}})\int\limits_0^t {{e^{{\lambda _1}(t - \tau )}}} u(\tau )d\tau \bullet {v_1} + \cdots + ({\alpha _{1k}} + {\alpha _{2k}})\int\limits_0^t {{e^{A(t - \tau )}}} u(\tau )d\tau \bullet {v_4}\\ = ({\alpha _{11}} + {\alpha _{21}})\int\limits_0^t {{e^{{\lambda _1}(t - \tau )}}} u(\tau )d\tau \bullet {v_1} + \cdots + ({\alpha _{1k}} + {\alpha _{2k}})\int\limits_0^t {{e^{{\lambda _k}(t - \tau )}}} u(\tau )d\tau \bullet {v_k}
0∫teA(t−τ)Bu(τ)dτ=(α11+α21)0∫teλ1(t−τ)u(τ)dτ∙v1+⋯+(α1k+α2k)0∫teA(t−τ)u(τ)dτ∙v4=(α11+α21)0∫teλ1(t−τ)u(τ)dτ∙v1+⋯+(α1k+α2k)0∫teλk(t−τ)u(τ)dτ∙vk
当
k
≥
4
k \ge 4
k≥4的时候,上式可以表示为
∫
0
t
e
A
(
t
−
τ
)
B
u
(
τ
)
d
τ
=
(
α
11
+
α
21
)
∫
0
t
e
λ
1
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
1
+
⋯
+
(
α
1
k
+
α
2
k
)
∫
0
t
e
A
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
4
=
(
α
11
+
α
21
)
∫
0
t
e
λ
1
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
1
+
⋯
+
(
α
14
+
α
24
)
∫
0
t
e
λ
4
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
4
+
δ
\int\limits_0^t {{e^{A(t - \tau )}}} Bu(\tau )d\tau \\ = ({\alpha _{11}} + {\alpha _{21}})\int\limits_0^t {{e^{{\lambda _1}(t - \tau )}}} u(\tau )d\tau \bullet {v_1} + \cdots + ({\alpha _{1k}} + {\alpha _{2k}})\int\limits_0^t {{e^{A(t - \tau )}}} u(\tau )d\tau \bullet {v_4}\\ = ({\alpha _{11}} + {\alpha _{21}})\int\limits_0^t {{e^{{\lambda _1}(t - \tau )}}} u(\tau )d\tau \bullet {v_1} + \cdots + ({\alpha _{14}} + {\alpha _{24}})\int\limits_0^t {{e^{{\lambda _4}(t - \tau )}}} u(\tau )d\tau \bullet {v_4} + \delta
0∫teA(t−τ)Bu(τ)dτ=(α11+α21)0∫teλ1(t−τ)u(τ)dτ∙v1+⋯+(α1k+α2k)0∫teA(t−τ)u(τ)dτ∙v4=(α11+α21)0∫teλ1(t−τ)u(τ)dτ∙v1+⋯+(α14+α24)0∫teλ4(t−τ)u(τ)dτ∙v4+δ
当
k
=
4
k = 4
k=4时,
δ
=
0
\delta = 0
δ=0
当
k
>
4
k > 4
k>4时,
δ
=
(
α
15
+
α
25
)
∫
0
t
e
A
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
5
⋯
+
(
α
1
k
+
α
2
k
)
∫
0
t
e
λ
4
(
t
−
τ
)
u
(
τ
)
d
τ
∙
v
4
\delta = ({\alpha _{15}} + {\alpha _{25}})\int\limits_0^t {{e^{A(t - \tau )}}} u(\tau )d\tau \bullet {v_5} \cdots + ({\alpha _{1k}} + {\alpha _{2k}})\int\limits_0^t {{e^{{\lambda _4}(t - \tau )}}} u(\tau )d\tau \bullet {v_4}
δ=(α15+α25)0∫teA(t−τ)u(τ)dτ∙v5⋯+(α1k+α2k)0∫teλ4(t−τ)u(τ)dτ∙v4
由上述分析可以得到当 k ≥ 4 k \ge 4 k≥4的时候,即 B B B的 的所有列属于一个存在的 A A A的真线性不变子空间的时候,系统完全能控,反之,不能包含 A A A中特征向量的所有方向,不能控。
标签:tau,粗浅,线性系统,matrix,limits,int,作业,cdots,alpha 来源: https://blog.csdn.net/weixin_40999869/article/details/114610131