Chapter 5

State Space Analysis

CHAPTER HIGHLIGHTS

☞ Basic Definitions ☞ Controllability

☞ State Space Representation ☞ Kalman’s Test for Controllability
☞ Transfer Function to State Space Model ☞ Observability
☞ Controllable Canonical Form ☞ Kalman’s Test for Observability
☞ Observable Canonical Form ☞ Solution of State Equations
☞ Diagonal Canonical Form ☞ Transfer Function
☞ Jordan Canonical Form

Basic Definitions D

System state Minimum information needed in order to x0

completely determine the output of the system from a given u(t ) x(t ) 1 x(t ) + y(t )
B + ∫= C +
moment provided the input is known from that moment. + S

System variable Any variable that responds to an input or

initial conditions in a system A
State variables The smallest set of linearly, independent
Figure 1 Block diagram representation of state space model
system variables such that the values of the set members
at time t0 along with known forcing function completely Advantages of State Space Analysis
determine the value of all system variables for all t ≥ t0.
State vector Vector whose elements are the state variables 1. It is applicable to multiple input and multiple output
State space n-dimensional space whose axes represent the systems.
state variables 2. It is applicable to systems with non-zero initial
State equations A set of n simultaneous, first-order dif- conditions.
ferential equations with n variables, where n variables to be 3. It is applicable to both linear time-invariant (LTI) and
solved are the state variables. non-linear time-varying systems.
State space representation 4. All the internal states can be determined.
A state space representation of an LTI system has the gen- 5. It is more accurate than transfer function (TF).
eral for x. 6. It gives the information about controllability and
•
observability.
x (t) = Ax(t) + Bu(t)
State space representation
y(t) = Cx(t) − Du(t) In order to select the state variables, we must follow the
x(t0) = X0 → initial conditions rules given here:
Where (a) A minimum number of state variables must be
x(t) : State vector (n dimensional) ­selected.
y(t) : Output vector (P dimensional) (b) They must be linearly independent.
u(t) : Input or control vector (m dimensional)
The minimum number of state variables required equals
A : Dynamic or system matrix (n × n)
the order of the differential equation describing the system.
B : Input matrix (n × m)
From the TF point of view, the order of the differential equa-
C : Output matrix (p × n)
tion is the order of the denominator of the TF after cancel-
D : Feed forward (direct) matrix (p × m)
ling common factors in the numerator and denominator.
 Chapter 5 • State Space Analysis | 3.327

A practical way to determine the number of state vari- Applying node equation at V1
ables is to count the number of independent energy stor-
1
age elements in the system (capacitor and inductors in iC = –iR + iL = − uc + iL (3)
electrical system and masses and springs in mechanical R
system). Applying loop equation in capacitor loop
The following is the procedure for state representation of
electrical network: VL = –VC + V(t)(4)
1. Write a simple derivative equation for each energy stor- From Eqs (1), (2), (3), and (4)
age element (node equation for the node at which an
inductor is connected and loop equation for the loop in dvC 1 1
=− uC + iL
which capacitor is connected in electrical network). dt RC C
2. Solve for the each derivate term as a linear combination diL 1 1
of the system variable and the input. = − u C + V (t )
dt L L
3. Each differentiated variable is selected as a state
1
variable. Output equation iR = uC
R
4. All other variables and output are written in terms of
the state variables and the input. Step 4: Obtain state space representation in vector matrix
form.

⎡ i ⎤ ⎡−
Solved Examples 1 1⎤
⎢ ⎡0⎤
u
⎢ C ⎥ = ⎢ RC C ⎥ ⎡uC ⎤ ⎢ ⎥
⎢i ⎥ ⎢ 1 ⎥ ⎢ ⎥ + 1 v (t )
Example 1 i ⎢ ⎥
⎣⎢ ⎦⎥ ⎢⎣ − L
L 0 ⎥ ⎣ L ⎦ ⎣2⎦
Find the state space representation of the system shown in ⎥⎦
the figure if the output is the current through the resistor.
⎤ ⎡VC ⎤
[iR ] = ⎡⎢ R
1
L 0⎥ ⎢ ⎥
⎣ ⎦ ⎣ iL ⎦
i(t )
V(t ) + R C
– Transfer Function To State
Space Model
This case corresponds to a linear system that can be repre-
Solution sented as an nth-order differential equation with constant
Step 1: Label all the branch variables in the network. coefficient as follows:
1 1
1 d n y(t ) d n −1 y(t ) d n − 2 y (t ) dy(t )
1 + −
an −1 + an − 2 + .... + a1 + a0
1 1 1 1 dt n dt n 1 dt n − 2 dt
− = boU(t)
−
The classical transfer function (TF) representation of the
system is obtained by applying the laplace transform to the
Step 2: Select the state variables. Write the derivative differential equation.
equations for all energy storage elements (L and C) Y (S ) bo
G(s) = = n
dv di U ( S ) s + an −1 s + an − 2 s n − 2 + .... + a1s + a0
n −1
C. c = ic; L. L = VL
dt dt
Choose the differentiated quantities as the state variables If the state space representation has to be obtained, a con-
(nC and iL). venient way to select state variables is to choose the output
y(t) and its ‘n – 1’ derivatives as the state variables. This is
Step 3: Write each differentiated term as a linear combina- called phase variables choice.
tion of system variables and the input. The state space representation using the phase variable
dvc choice of the state variables is said to be in the controllable
C. = iC = f1(nC, iC, V(t)) (1)
dt canonical form (CCF).
diL
L. = uc = f2(nC, iL, V(t))  (2)
dt
3.328 | Part III • Unit 3 • Control Systems

Controller canonical form of the given differential equation is as follows:

⎡ • ⎤
⎢ x1 ⎥
⎢ • ⎥
⎢ x2 ⎥ ⎡ 0 1 0 0 . . . . 0 ⎤ ⎡ x1 ⎤ ⎡ 0 ⎤ ⎡ x1 ⎤
⎢ • ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢x ⎥
⎢ x3 ⎥ ⎢ 0 0 1 0 . . . . 0 ⎥ ⎢ x 2 ⎥ ⎢ ⎥0 ⎢ 2⎥
⎢ ⎥ ⎢ 0 ⎢ x3 ⎥
0 0 1 . . . . 0 ⎥ ⎢ x3 ⎥ ⎢ 0 ⎥ ⎢ ⎥
⎢ • ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ U .
⎢ x 4 ⎥ ⎢ 0 0 0 0 1 . . . 0 x
⎥ ⎢ 4 ⎥ ⎢ ⎥ 0 Y = [1 0 0 . . . . 0] ⎢⎢ . ⎥⎥
⎢ . ⎥=⎢ . ⎥ ⎢ . ⎥+⎢ . ⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ . ⎥ ⎢ . ⎥ ⎢ . ⎥ ⎢ ⎥. ⎢.⎥
⎢ . ⎥ ⎢ . ⎥ ⎢ . ⎥ ⎢.⎥ ⎢.⎥
⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ . ⎥ ⎢ 0 0 0 0 0 0 . . 1 ⎥ ⎢ xn −1 ⎥ ⎢ . ⎥ ⎢⎣ xn ⎥⎦
⎢• ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ x n −1 ⎥ ⎣ − a0 − a1 − a2 − a3 . . . . − an −1 ⎦ ⎣ xn ⎦ ⎣bo ⎦
⎢ ⎥
⎢ • ⎥
⎣ xn ⎦
Consider a transfer function with polynomial in numerator: Y ( S ) b0 s n + b1s n −1 + ....bn −1s + bn
=
y + a1y
n n – 1 + a2yn – 2 + ……+ an–1 y + any = bo u + b1u
• n n–1
U ( S ) s n + a1s n −1 + ... + an −1s + an
•
+ …..+ bn–1 u State space representation of the system in controllable
+ bnu. canonical form and observable canonical form are given as
where U is the input and y is the output. The transfer func- follows:
tion can be written as follows:

Controllable Canonical Form

⎡ • ⎤
⎢ x1 ⎥
⎢ • ⎥ ⎡ 0 1 0 . . . . . 0 ⎤ ⎡ x1 ⎤ ⎡0 ⎤
⎢ x2 ⎥ ⎢
⎢ ⎥ ⎢ 0 0 1 . . . . . 0 ⎥⎥ ⎢⎢ x2 ⎥⎥ ⎢⎢0 ⎥⎥
⎢ . ⎥ ⎢ . . . . . . . . . ⎥ ⎢ . ⎥ ⎢.⎥ ⎡ x1 ⎤
⎢ . ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢x ⎥
⎢ ⎥ ⎢ . . . . . . . . . ⎥ ⎢ . ⎥ ⎢.⎥
⎢ 2⎥
⎢ . ⎥=⎢ ⎥ ⎢ . ⎥ + ⎢ . ⎥ U y = [b – a b :b - a b :……:b – a b ] ⎢ . ⎥ + b u.
⎢ . ⎥ ⎢ . . . . . . . . . n n o n–1 n–1 o 1 1 0 o
⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ . . . . . . . . . ⎥ ⎢ . ⎥ ⎢.⎥ ⎢ .⎥
⎢ . ⎥ ⎢
⎢ . ⎥ ⎢ . . . . . . . . . ⎥ ⎢ . ⎥ ⎢.⎥ ⎢.⎥
⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ 0 0 0 . . . . . 1 ⎥ ⎢ xn −1 ⎥ ⎢0 ⎥ ⎢.⎥
⎢ • ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢.⎥
⎢ x n −1 ⎥ ⎣ − an − an −1 − an − 2 . . . . . − a1 ⎦ ⎣ xn ⎦ ⎣1 ⎦ ⎢ ⎥
⎢ • ⎥ ⎢⎣ xn ⎥⎦
⎢⎣ x n ⎥⎦
The controllable canonical form is important in discussing Observable Canonical Form
the pole placement approach to the control system design. The following state–space representation is called an ob-
servable canonical form.
⎡ ⎤
•
x
⎢ ⎥ ⎡0 0 . . . . . .0 − a
1
n ⎤ ⎡ x1 ⎤ ⎡bn − an b0 ⎤ ⎡ x1 ⎤
⎢• ⎥ ⎢ ⎥
⎢ x 2 ⎥ ⎢1 0 . . . . . . 0 − an −1 ⎥ ⎢⎢ x2 ⎥⎥ ⎢⎢bn −1 − an −1b0 ⎥⎥ ⎢ x ⎥
⎢• ⎥ ⎢ ⎥ ⎢. ⎥ ⎢. ⎢ 2 ⎥
⎢ x 3 ⎥ ⎢. ⎥ ⎢ . ⎥
⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢. ⎥ .
⎢ ⎥+⎢ . ⎥u
⎢. ⎥ = ⎢ ⎢ . ⎥
⎥ ⎢.. ⎥ ⎢. ⎥ y = [1 0 0 . . . .0 1] +b u
⎢. ⎥ ⎢. ⎥⎢ ⎥ ⎢ ⎢ . ⎥ 0
⎢ ⎥ ⎢. ⎥ ⎢ ⎥
⎥ ⎢. ⎥ ⎢. ⎥ ⎢ . ⎥
⎢. ⎥ ⎢ ⎥ ⎢. ⎥ ⎢. ⎥
⎢ • ⎥ ⎢. ⎥⎢ ⎥ ⎢ ⎢. x ⎥
⎢x ⎥ ⎢ ⎥ ⎢ n −1 ⎥
⎥⎦ ⎣⎢ xn ⎦⎥ ⎣⎢b1 − a1 , b0 ⎦⎥
⎢ n ⎥ ⎣0 0 . . . . . 1 − a1 ⎣⎢ xn ⎦⎥
⎢⎣ ⎥⎦
 Chapter 5 • State Space Analysis | 3.329

System matrix in observable canonical form is the trans- C1 C2 C

pose of system matrix in controllable canonical form. = b0 + + + ... n
s + p1 s + P2 s + Pn
Diagonal Canonical Form where p1, p2, … pn are the locations of poles.
Transfer function with numerator polynomial can be writ- The diagonal canonical form of the state space representa-
ten as follows: tion of this system is given as follows:

Y ( s) b0 s n + b1s n −1 + ...bn −1s + bn

=
U ( s) ( s + p1 )( s + p2 )( s + p3 )...( s + pn )

⎡ x1 ⎤ ⎡ − p1 . . . . . . . 0 ⎤ ⎡ x1 ⎤ ⎡1⎤ ⎡ x1 ⎤
⎢x ⎥ ⎢ 0 − p2 . . . . . . 0 ⎥⎥ ⎢⎢ x2 ⎥⎥ ⎢⎢1⎥⎥ ⎢x ⎥
⎢ 2⎥ ⎢ ⎢ 2⎥
⎢.⎥ ⎢ . ⎥ ⎢ . ⎥ ⎢.⎥ ⎢.⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢.⎥ ⎢ . ⎥ ⎢ . ⎥ ⎢ . ⎥ u y = [C1 C2 . . . .Cn] ⎢ . ⎥ + bou
⎢ . ⎥=⎢ . ⎥ ⎢ . ⎥ + ⎢.⎥ ⎢.⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢.⎥ ⎢ . ⎥ ⎢ . ⎥ ⎢.⎥ ⎢.⎥
⎢.⎥ ⎢ . ⎥ ⎢ . ⎥ ⎢..⎥ ⎢ .. ⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢.⎥ ⎢ . ⎥ ⎢ . ⎥ ⎢.⎥ ⎢.⎥
⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎣ xn ⎦ ⎣ 0 − pn ⎦ ⎣ xn ⎦ ⎣1⎦ ⎣ xn ⎦

Jordan Canonical Form

When the system involves multiple roots, the diagonal For example, if there are three equal poles (p1= P2 = P3), the
canonical form must be modified into Jordon canonical form. factored form Y(s)/U(s) becomes

Y ( s) C1 C2 C3 C4 Cn
= b0 + + + + + ... +
U ( s) ( s + P1 ) 3 ( s + p1 ) 2 ( s + p1 ) s + P4 s + pn
⎡• ⎤
⎢ x1 ⎥ ⎡ − p1 1 0 1 0 . . . 0 ⎤
⎢• ⎥ ⎢
0 − p 1 1 . 0 . . . ⎥ ⎡ x1 ⎤ ⎡0 ⎤ ⎡ x1 ⎤
⎢ x2 ⎥ ⎢ 1 ⎥ ⎢ x2 ⎥ ⎢ 0 ⎥ ⎢x ⎥
⎢• ⎥ ⎢ 0 0 − p1 1 0 . . . 0 ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ 2⎥
⎢ x 3 .⎥ ⎢ ⎥ ⎢ x3 .⎥ ⎢1 ⎥ ⎢.⎥
⎢ ⎥ ⎢ − − − − − − − . ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
. x4 1 u y = [C C . . . . .C ]
⎢ • ⎥ ⎢ 0 0 0 . − p4 . . . 0 ⎥ ⎢ ⎥ ⎢ ⎥ 1 2 n ⎢ . ⎥ + bou.
⎢. x 4 ⎥ = ⎢ ⎥ ⎢ . ⎥ + ⎢.⎥ ⎢.⎥
⎢ . ⎥ ⎢ . . ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ . . . ⎥ ⎢ . ⎥ ⎢.⎥ ⎢.⎥
⎢ . ⎥ ⎢ ⎥ ⎢ . ⎥ ⎢ ..⎥ ⎢ .. ⎥
⎢ . ⎥ ⎢ . . ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
⎢ ⎥ ⎢ ⎥ ⎢ . ⎥ ⎢.⎥ ⎢.⎥
⎢ . ⎥ ⎢ . . . ⎥ ⎢ x ⎥ ⎢1 ⎥ ⎢ ⎥
⎢ • ⎥ ⎢⎣ 0 ⎣ n⎦ ⎣ ⎦ ⎣ xn ⎦
0 0 . 0 . . . − pn ⎥⎦
⎢ xn ⎥
⎣ ⎦

Eigen values of an n × n matrix A NOTES

The characteristic equation of a square matrix ‘A’ is given 1. The poles of the transfer function are given by the
as follows: sI − A = 0. This function is the same equation as the
lI −A =0 characteristic equation of A. Therefore, we can con-
clude that the eigen values of the state model and the
The values of l which satisfy the characteristic equation are poles of the transfer function are the same.
called eigen values of A matrix. 2. Stability of the SISO (single input single output) sys-
tem depends on the eigen values of system matrix in
its state space model.
3.330 | Part III • Unit 3 • Control Systems

Controllability The solution for the above state equation is

t
A state x(t) is said to be controllable at t = t0 if there exists
a piecewise continuous input u(t) that will drive the state to x(t ) = e At x(0) + ∫ e − A( t −t ) Bu(t ) . dt
any final state x(tf) for a finite time (tf − t0) ≥ 0. Homogeneous 0 Forced Solution
If every state x(t0) of the system is controllable in a finite Solution
time interval, the system is said to be completely control- In the absence of the input to the system, the response of the
lable or simply, controllable. system or solution of state equations with initial conditions
alone is given as follows:
Kalman’s Test for Controllability
Consider a state space model: X(t) e At x(0) = e At x0
• It is observed that the initial state x0 at t = 0 is driven to a
x = Ax + Bu state x(t) at time ‘t’. This transition in state is carried out
Then, controllability matrix by the matrix exponential eAt. Due to this property, eAt is
Qc = [B:AB:A2B : ….:An–1 B] known as state transition matrix and is denoted by φ(t).

Properties of State Transition Matrix f (t)

NOTES The following are the properties of state transition matrix:
1. 
A system is said to be controllable if the rank of QC is 1. f (0) = eA0 = I
equal to the order of matrix A or Qc ≠ 0.
2. f (t) = eAt = (e–At)–1 = [f (–t)]–1
2. 
The number of uncontrollable states is computed as
3. f (t1 + t2) = e A( t1 + t2 ) = e At1 . e At2
the difference between order of A matrix (n) and rank
   = f (t1). f (t2) = f (t2). f (t1)
of QC matrix (P).
4. [f (t)]n = [eAt]n = eAn+ = f (nt)
Observability 5. f (t1 – t) . f (t − t2) = eA(t1 – t) . eA(t – t2)
    = eA(t1 - t2) = f (t1 – t2)
An LTI system is said to be observable if in given any input
u(t), there exists a finite time tf ≥ t0 such that the knowledge Transfer Function
of u(t) for t0 ≤ t, < ff, matrix A, B, C, and D and the output Given the state space model of SISO system as
y(t) for t0 ≤ t < tf are sufficient to determined x(t0). •
If every state of the system is observable for a finite tf, x = Ax + Bu
we say that the system is completely observable or simply, Y = Cx + Du
observable. The transfer function of the system is

Kalman’s Test for Observability TF = C[sI – A]–1 B + D

Consider a state space model: Example 2
•
x = Ax + Bu and y = Cx + Du The maximum number of states required to describe the
Then observability matrix network as shown in figure is
R1 R2
Q0 = [C CA CA2 ……CAn–1]T
R3 R4
NOTES C
1. The system is said to be observable if the rank of Q0 is L
equal to the order of matrix A
or Qc ≠ 0. (A) 1 (B) 2 (C) 3 (D) 4
2. 
The number of unobservable states is computed as the
Solution
difference between order of A matrix (n) and rank of Q0
No. of energy storage elements (L, C) = 2
matrix (q).
No. of states required to analysis = 2

Solution of State Equations Example 3

The state equation of a linear system is as follows: The matrix of any state space equation for the transfer
• C ( s)
x = Ax + Bu; x(t0) = x(0) : function of the system shown in figure is
R( s)
 Chapter 5 • State Space Analysis | 3.331

R(S )
⎡ s + 1 −1 ⎤ ⎡1⎤
4 +
1 C(s)
[0 1] ⎢ s + 4 ⎥⎦ ⎢1⎥
– s
TF = −C[sI − A]–1 B = ⎣ 3 ⎣⎦
s 2 + 5s + 7
⎡ s +1−1 ⎤
[0 1] ⎢3 + s + 4⎥
⎡ −1 0 ⎤ = ⎣ ⎦
(A) ⎢ ⎥ (B) [ −1] s 2 + 5s + 7
⎣ 0 1⎦
⎡ −1 0 ⎤ s+7
(C) [4] (D) ⎢ =
⎥ s2 + 5s + 7
⎣ 0 4⎦
Example 5
Solution Given the homogeneous state space equations
Output of integrator is considered as state and number of • ⎡ −4 1 ⎤
integrators is equal to the number of states. x = ⎢ ⎥ x, the steady-state value of xss = t→∞
Lt x(t),
⎣ 0 −3⎦
Number of states = 1
given the initial state value of x[0] = [10 – 10]T is
R(S ) x1 1
x1 C(S) ⎡ 10 ⎤ ⎡ −4 ⎤
4 + (A) xss = ⎢ ⎥ (B) xss = ⎢ ⎥
– S ⎣ −10 ⎦ ⎣ −3⎦
⎡0 ⎤ ⎡∞ ⎤
(C) xss = ⎢ ⎥ (D) xss = ⎢ ⎥
⎣0 ⎦ ⎣∞ ⎦
•
x 1 = –x1 + 4r(t)
Solution
[x°1] = [–1] [x1] + [4] [r(t)]
Solution of homogenous equation
Matrix of state space equation is [–1].
•
A x = Ax
Example 4
x(t) = eAt x(0)
A system is described by the state equations as follows:
• eAt = L ⎡⎣[ sI − A]−1 ⎤⎦
x = Ax + Bu . the output is given by y = Cx
⎡ −4 −1⎤ ⎡1⎤ s + 4 −1 ⎤
where A = ⎢ ⎥;B= ⎢1⎥ ; C = [0 1] sI – A = ⎢⎡
⎣ 3 −1⎦ ⎣⎦ ⎣ 0 s + 3⎥⎦
The transfer function G(s) of the system is −1 ⎡s + 3 1 ⎤
[ sI − A]−1 = ⎢
(A) s (B) s+7 ( s + 4)( s + 3) ⎣ 0 s + 4 ⎥⎦
s + 5s + 7
2 s + 5s + 7
2
⎡ 1 1 ⎤
1 s ⎢s + 4
(C) (D) ( s + 4)( s + 3) ⎥
s 2 + 5s + 7 s2 + s + 5 = ⎢ ⎥
⎢ 1 ⎥
Solution ⎢⎣ 0 s+3 ⎥⎦
Transfer function of the state space model is
⎡e −4t e −3t − e −4t ⎤
Y ( s) eAt = L−1[( sI − A) −1 ] = ⎢ ⎥
TF = = C[sI – A]–1B + D ⎣ 0 e −3t ⎦
U ( s)
⎡10e −4t + 10e −3t + 10e −4t ⎤
⎡1 0 ⎤ ⎡ −4 −1⎤ x(t) = e At x(0) = ⎢ ⎥
[sI − A] = s ⎢ ⎥−⎢ ⎥ ⎣ −10e −3t ⎦
⎣0 1 ⎦ ⎣ 3 −1⎦
⎡ 20e −4t − 10e −3t ⎤
⎡s + 4 1 ⎤ x(t) = ⎢ ⎥
= ⎢ ⎥ ⎣ −10 −3t ⎦
⎣ −3 s + 1⎦
⎡ lt ( 20e −4t − 10e −3t ) ⎤
1 ⎡s + 1 − 1 ⎤ xss = lt x(t ) = ⎢t →∞ ⎥ = ⎡0 ⎤
[sI – A]–1 = ⎢ lt − 10e −3t ⎥ ⎢⎣0 ⎥⎦
( s + 1)( s + 4) ⎢⎣3 s + 4 ⎥⎦
t →∞
⎣ t →∞ ⎦
3.332 | Part III • Unit 3 • Control Systems

Example 6 Solution
A second-order system starts with an initial condition of From the standard controllable canonical form of the trans-
⎡ 2⎤ fer function
⎢ 3⎥ without any external input. The state transition matrix 2s + 4
⎣ ⎦ TF = 3
⎡e −3t 0 ⎤ s + 4 s2 + 9s + 4
for the system is given by ⎢ − ⎥
⎣ 0 e 4t ⎦
State space representation
The state of the system at end of 2 s is given by
• ⎡0 1 0⎤ ⎡ x1 ⎤ ⎡0 ⎤
⎡1 ⎤ ⎡ 5⎤ X (t ) = ⎢ 0
(A) 10 −3 ×⎢ ⎥ (B) 10 −3 ×⎢ ⎥ ⎢ 0 1 ⎥⎥ ⎢ x ⎥ + ⎢0 ⎥ u(t)
⎢ 2⎥ ⎢ ⎥
⎣ 5⎦ ⎣1 ⎦ ⎢⎣ −4 −9 −4 ⎥⎦ ⎢⎣ x3 ⎥⎦ ⎢⎣ 2⎥⎦
⎡1 ⎤ ⎡ 5⎤
(C) ⎢ ⎥ (D) ⎢ ⎥
⎣ 5⎦ ⎣1 ⎦ Example 8
• ⎡3 4⎤ ⎡1 ⎤
Solution For the system x = ⎢
0 5 ⎥ x + ⎢0 ⎥ u, which of the following
⎣ ⎦ ⎣ ⎦
Solution of state equations without external input is
statements about the system is true?
x(t) = eAt × (0) (A) Controllable and stable
(B) Uncontrollable and stable
⎡e −3t 0 ⎤ ⎡ 2⎤
= ⎢ ⎥⎢ ⎥ (C) Controllable and unstable
⎣ 0 e −4t 3
⎦⎣ ⎦ (D) Uncontrollable and unstable
⎡ 2e −3t ⎤ Solution
X(t) = ⎢ −4t ⎥
⎣3e ⎦ For stability analysis, location of poles are the roots of char-
acteristic equation.
⎡ 2e −3× 2 ⎤ ⎡5⎤
At t = 2 s, x(2) = ⎢ −4 × 2 ⎥ = ⎢ ⎥ .10 −3 s − 3 −4
⎣3e ⎦ ⎣1 ⎦ sI − A = = (s – 3) (s – 5) = 0
0 s−5
Example 7 s = 3 and 5
For a system with transfer function Therefore, poles are located on RHS plane, system is
G(s) = 2s + 4 unstable.
s + 4 s2 + 9s + 4
3 Controllability matrix QC = [ B AB ]
•
The matrix A in the state space form x = Ax + Bu is equal ⎡ 3 4 ⎤ ⎡1 ⎤ ⎡ 3⎤
[A B] = ⎢ ⎥ ⎢ ⎥=⎢ ⎥
to ⎣0 5 ⎦ ⎣0 ⎦ ⎣0 ⎦
⎡0 1 0⎤ ⎡1 0 0⎤ ⎡1 3⎤
⎢
QC = ⎢ ⎥ ⇒ QC = 0
(A) ⎢ 0 0 1 ⎥⎥ ⎢
(B) ⎢ 0 1 0 ⎥⎥ ⎣0 0 ⎦
⎢⎣ −4 −9 −4 ⎥⎦ ⎢⎣ −4 −9 −4 ⎥⎦ Therefore, the system is uncontrollable.

⎡1 0 0⎤ ⎡0 1 0⎤
(C) ⎢⎢ 0 0 1 ⎥⎥ (D) ⎢⎢ −2 −4 1 ⎥⎥
⎢⎣ −2 −4 −1⎥⎦ ⎢⎣ −4 −9 −4 ⎥⎦

Exercises
Practice Problems 1
⎡1 0⎤ ⎡1 0⎤
Direction for questions 1 to 15: Select the correct alterna- (A) ⎢ ⎥ (B) ⎢ ⎥
tive from the given choices. ⎣ −2 −1⎦ ⎣ −1 −3⎦
•
⎡0 1⎤ ⎡0 1⎤
1. Given that X = AX for the system described by the dif- (C) ⎢ ⎥ (D) ⎢ ⎥
•• • ⎣ −3 −1⎦ ⎣ −3 −2⎦
ferential equation y + 2 y + 3 y = 0 . The matrix A is
 Chapter 5 • State Space Analysis | 3.333

2. A signal flow graph of a system is given below: The state transition matrix f(t) is

–α
⎡e −3t 0⎤ ⎡e −3t 0 ⎤
(A) ⎢ ⎥ (B) ⎢ ⎥
⎣ 0 e3t ⎦ ⎣ 0 e −3t⎦
1
1/s
x1 ⎡e −t / 30 ⎤ 0 ⎤
⎡et / 3
U1 α
(C) ⎢ t / 3 ⎥ (D) ⎢ − t /3 ⎥
−β ⎣ 0 e ⎦ ⎣ 0 e ⎦
β x3 X3
1
5. A particular control system is described by the follow-
1 1/s γ s ing state equations:
U2 ⎡ 0 1 ⎤ ⎡ x1 ⎤ ⎡0 ⎤
x2 X =⎢ ⎥⎢ ⎥ + ⎢ ⎥ U and
⎣ −5 −2⎦ ⎣ x2 ⎦ ⎣1 ⎦
⎡x ⎤
– γ Y = [1 0] ⎢ 1 ⎥
⎣ x2 ⎦
If the state equation that corresponds to the above sig- The transfer function of this system is
nal flow graph is X = AX + BU, the matrix A and B Y ( s) 2 Y ( s) 4
are (A) = (B) =
U ( s) s 2 + 2 s + 5 U ( s) s 2 + 2 s + 5
⎡b −g 0⎤ ⎡0 0 ⎤
Y ( s) 1 Y ( s) 3
(A) ⎢⎢ g a 0 ⎥ and ⎢0 1 ⎥⎥
⎥ ⎢ (C) = (D) =
U ( s) s 2 + 2 s + 5 U ( s) s 2 + 2 s + 5
⎢⎣ −a −b 0 ⎥⎦ ⎢⎣1 0 ⎥⎦
Direction for questions 6 and 7:
⎡0 a g ⎤ ⎡1 0 ⎤ A system is characterized by the following state space
(B) ⎢⎢0 −a −g ⎥⎥ and ⎢⎢0 1 ⎥⎥ equations
⎢⎣0 b − b ⎥⎦ ⎢⎣0 0 ⎥⎦ ⎡ x• ⎤
⎢ 1 ⎥ = ⎡ −3 1 ⎤ ⎡ x1 ⎤ + ⎡0 ⎤ U , (t > 0)
⎢ • ⎥ ⎢⎣ −2 0 ⎥⎦ ⎢⎣ x2 ⎥⎦ ⎢⎣1 ⎥⎦
⎡ −a b 0⎤ ⎡1 0 ⎤ ⎢⎣ x2 ⎥⎦
⎢
(C) ⎢ − b −g 0 ⎥ and ⎢⎢0 1 ⎥⎥
⎥
⎡ x1 ⎤
⎢⎣ a g 0 ⎥⎦ ⎢⎣0 0 ⎥⎦ Y = [1 0] ⎢ ⎥
⎣ x2 ⎦
⎡ −a 0 b ⎤ ⎡0 0 ⎤ 6. The transfer function of the system is
(D) ⎢⎢ g 0 a ⎥⎥ and ⎢⎢0 1 ⎥⎥ (A)
s
(B)
1
⎢⎣ − b 0 −a ⎥⎦ ⎢⎣1 0 ⎥⎦ ( s + 2) ( s + 1) s ( s + 2) ( s + 1)
s 1
(C) (D)
3. U(t ) x2 s –1 s –1 ( s − 2) ( s + 1) ( s + 2) ( s + 1)
x1 1
x2 x1 7. The state transition matrix of the system is
Y
–1 –2 ⎡ e − t + 2e −2t e − t + e −2t ⎤
(A) ⎢ −t − 2t ⎥
1 ⎣ −2e + 2 e 2e − t + e −2t ⎦
In the state diagram of a system shown in the figure, ⎡ − e − t + 2e −2t e − t − e −2t ⎤
which variables are controllable? (B) ⎢ − − ⎥
⎣ −2e + 2e
t 2 t 2e − t − e −2t ⎦
(A) x1(t)
(B) x2(t) ⎡ e − t − 2e −2t e − t − e −2t ⎤
(C) ⎢ −t −2t ⎥
⎣ −2e − 2e 2e − t − e −2t ⎦
(C) Both x1(t) and x2(t)
⎡ e − t + 2e −2t e − t − e −2t ⎤
(D) Neither x1(t) nor x2(t) (D) ⎢ ⎥
−t −2t
⎣ −2e − 2e 2e − t + e −2t ⎦
4. The state equations of a linear time-invariant system
8. Match List–I (Matrix) with List–II (dimensions) for the
dx(t )
are represented by = AX (t ) + BU (t ) state equations:
dt •
X (t ) = AΧ(t ) + BU (t ) and Y(t) = CX(t) + DU(t) and
⎡ −3 0 ⎤ ⎡0 ⎤
A= ⎢ ⎥ B= ⎢ ⎥ select the correct answer using the codes given in the
⎣ 0 −3⎦ ⎣1 ⎦ lists:
3.334 | Part III • Unit 3 • Control Systems

List-I List-II Which of the following statements about the control-

A
lability of the system are true?
(1) n × p
(A) The system is completely controllable for any non-
B (2) q × n
zero values of p and q.
C (3) n × n (B) only p = 0 and q = 0 result in controllability.
D (4) q × p (C) The system is uncontrollable for all values of
Codes: p and q.
(D) we cannot conclude about controllability from the
A B C D
given data.
(A) 1 3 4 2
(B) 1 3 2 4 13. The state transition matrix of the system whose state
(C) 3 1 4 2 ⎡. ⎤
x1 ⎡ 0 1⎤ ⎡ x1 ⎤
(D) 3 1 2 4 equation ⎢ ⎥ = ⎢ ⎢ x ⎥ is
⎢ . ⎥ ⎣ −2 0 ⎥⎦ ⎣ 2⎦
9. Consider the single input, single output system with its ⎢⎣ x 2 ⎥⎦
state variable representation:
⎡ −1 0 0 ⎤ ⎡1 ⎤ ⎡ 1 ⎤
•
⎢ cos 2t cos 2t ⎥
X = ⎢ 0 −2 0 ⎥ X + ⎢1 ⎥⎥ U, Y = [1 0 2] X
⎢ ⎥ ⎢ (A) ⎢ 2 ⎥
⎢⎣ 0 0 −3⎥⎦ ⎢⎣0 ⎥⎦ ⎢⎣ − 2 sin 2t sin 2t ⎥⎦
The system is ⎡ 1 ⎤
(A) neither controllable nor observable. cos 2t sin 2t ⎥
(B) ⎢ 2
(B) controllable but not observable. ⎢ ⎥
(C) uncontrollable but observable. ⎢⎣ − 2 sin 2t cos 2t ⎥⎦
(D) both controllable and observable.
⎡ sin t − cos t ⎤
⎡ et 0 ⎤ (C) ⎢
10. Consider the state transition matrix: f(t) = ⎢ t t⎥
. ⎣ + cos t sin t ⎥⎦
⎣te e ⎦
The eigen values of the matrix when t = 0 are
⎡ sin 2t cos 2t ⎤
(A) 1, 2 (B) 2, 1 (D) ⎢ ⎥
(C) 1, 1 (D) 1, 3 ⎢⎣ − cos 2t sin 2t ⎥⎦
11. A linear system
• is described by the following state
14. If the eigen values of a 3 × 3 matrix A are 1, – 3, and 4.
equation - X = AX(t) + BU(t).
The eigen values of P-1AP (∴ P is a linear transforma-
⎡ 0 1⎤ tion) are
A = ⎢ ⎥. The state transition matrix of the system is
⎣ −1 0 ⎦ −1 1
(A) 1, , and (B) 1, – 3, and 4
3 4
⎡ cos t sin t ⎤ ⎡ − cos t sin t ⎤
(A) ⎢ ⎥ (B) ⎢ ⎥ (C) 1, 9, and 16 (D) –1, 3, and – 4
⎣ − sin t cos t ⎦ ⎣ − sin t − cos t ⎦
⎡ − cos t − sin t ⎤ ⎡cos t − sin t ⎤ ⎡0 1⎤
(C) ⎢ (D) ⎢ 15. The state equations of a system are x = ⎢ ⎥
⎥ ⎥ ⎣ −1 −2⎦
⎣ − sin t cos t ⎦ ⎣cos t sin t ⎦ ⎡0 ⎤
x + ⎢ ⎥ u. The closed-loop poles of the system are
dx ⎡1 0 ⎤ ⎣1 ⎦
12. Consider the system = AX + BU with A = ⎢0 1⎥
dt ⎣ ⎦ (A) ±2 (B) – 1, – 1
⎡ p⎤ (C) + 1, – 2 (D) + 2, – 1
and B = ⎢ ⎥ where p and q are arbitrary real numbers.
⎣q⎦

Practice Problems 2
⎡e −t e − t + e −2t ⎤ ⎡e −t e − t − e −2t ⎤
Direction for questions 1 to 15: Select the correct alterna- (A) ⎢ ⎥ (B) ⎢ ⎥
tive from the given choices. ⎣ 0 e 2t ⎦ ⎣ 0 e −2t ⎦
1. The state equation of a linear system is given by ⎡ 0 e −2t ⎤ ⎡ 0 e − t − e −2t ⎤
(C) ⎢ − t ⎥ (D) ⎢ − t ⎥
⎡ −1 1 ⎤ ⎡0 ⎤ ⎣e e − t − e −2t ⎦ ⎣e e −2t ⎦
X = AX + BU, where A = ⎢ ⎥ and B = ⎢ ⎥ . The
⎣ 0 −2⎦ ⎣1 ⎦
state transition matrix of the system is
 Chapter 5 • State Space Analysis | 3.335

2. The eigen value and eigen vector pairs (lI, Vi) for the ⎡ ⎤
X1 ⎡106 103 ⎤ ⎡ x1 ⎤ ⎡ 0 ⎤
system are (A) ⎢ ⎥ = ⎢ 9 ⎥ ⎢ ⎥ + ⎢ 3 ⎥ × 10
⎢. ⎥ 10 0 ⎦ ⎣ x2 ⎦ ⎣10 ⎦
⎡ ⎡ 1 ⎤⎤ ⎡ ⎡ 1 ⎤⎤ ⎢⎣ X 2 ⎥⎦ ⎣
(A) ⎢ −1, ⎢ ⎥ ⎥ , and ⎢ −2, ⎢ ⎥ ⎥
⎣ ⎣ −1⎦ ⎦ ⎣ ⎣ −2⎦ ⎦ ⎡ X1 ⎤ ⎡109 0 ⎤ ⎡ x ⎤ ⎡ 0 ⎤
1
⎡ (B) ⎢ . ⎥ = ⎢ 6 3 ⎥ ⎢x ⎥
+ ⎢ 3 ⎥ × 10
⎡ 1 ⎤⎤ ⎡ ⎡ 1 ⎤⎤ ⎢⎣ X 2 ⎥⎦ ⎣10 10 ⎦ ⎣ 2 ⎦ ⎣10 ⎦
(B) ⎢ −2, ⎢ ⎥ ⎥ , and ⎢ −1, ⎢ ⎥ ⎥
⎣ ⎣ −1⎦ ⎦ ⎣ ⎣ −2⎦ ⎦
⎡ ⎤
X1 ⎡ −106 −103 ⎤ ⎡ x1 ⎤ ⎡103 ⎤
⎡ ⎡ 1 ⎤⎤ ⎡ ⎡ 1 ⎤⎤ (C) ⎢ ⎥ = ⎢ 9 ⎥ ⎢ x ⎥ + ⎢ ⎥ × 10
(C) ⎢ −1, ⎢ ⎥ ⎥ , and ⎢ 2, ⎢ ⎥ ⎥ ⎢. ⎥
⎣ −1⎦ ⎦ ⎣ 10 0 ⎦ ⎣ 2⎦ ⎣ 0 ⎦
⎣ ⎣ ⎣ −2⎦ ⎦ ⎣⎢ X 2 ⎥⎦
⎡ ⎡ 1 ⎤⎤ ⎡ ⎡ 1 ⎤⎤ (D) None of these
(D) ⎢ 2, ⎢ ⎥ ⎥ , and ⎢1, ⎢ ⎥ ⎥ 7. The poles and zero of the following system
⎣ ⎣ −1⎦ ⎦ ⎣ ⎣ −2⎦ ⎦
⎡ 0 1⎤ ⎡0 ⎤
3. The system matrix A is X = ⎢ ⎥ x+ ⎢ ⎥u
⎣ −20 −9⎦ ⎣1 ⎦
⎡ 0 1⎤ ⎡1 1⎤
(A) ⎢ ⎥ (B) ⎢ ⎥ y = [–17 – 5]x + [1]u are
⎣ −1 1⎦ ⎣ −1 −2⎦
(A) poles: – 4, – 5, and zeros: – 1, – 3
⎡2 1⎤ ⎡0 1⎤ (B) poles: – 3, – 4, and zeros: – 3, – 2
(C) ⎢ ⎥ (D) ⎢ ⎥
⎣ −1 −1⎦ ⎣ −2 −3⎦ (C) poles: – 2, – 3, and zeros: – 2, – 4
(D) None of these
• ⎡1 2⎤ ⎡0⎤ . ⎡ −1 −2⎤
4. Let X = X = ⎢ X + ⎢ ⎥ U , Y = [b 0 ] X 8. The second-order system x = Ax has A = ⎢ ⎥.
⎥ ⎣ +1 −1⎦
⎣0 1⎦ ⎣1⎦
The values of its damping factor and natural frequency
where b is an unknown constant. This system is wn are, respectively
(A) observable for all values of b. (A) 1.732, 0.577 (B) 1.414, 0.6
(B) unobservable for all values of b. (C) 0.577, 1.732 (D) 0.6, 1.414
(C) observable for all non-zero values of b.
(D) unobservable for all non-zero values of b. Direction for questions 9 and 10:
5. A state variable representation of a system is given by Consider the following block diagram
the expression + +
U(s) 1/s 1/s 3
– +
⎡ ⎤ Y(s)
⎢ x1 ⎥ = ⎡1 0 ⎤ ⎡ x1 ⎤ + ⎡0 ⎤ U (t ) ,
⎢ • ⎥ ⎢⎣0 −1⎥⎦ ⎢⎣ x2 ⎥⎦ ⎢⎣1 ⎥⎦ 2
⎢⎣ x2 ⎥⎦ 7
⎡ x1 ⎤ 9. In the above figure, the state space representation in the
Y = [1 1] ⎢ ⎥
⎣ x2 ⎦ vector matrix from is
The transfer function of the system is ⎡ 0 1⎤ ⎡0 ⎤
(A) x = ⎢ ⎥ x + ⎢1 ⎥ u
2 2s ⎣ −7 2 ⎦ ⎣ ⎦
(A) (B)
s +1 ( s − 1)( s + 1) Y = [3 1] x
2 1 ⎡0 1⎤ ⎡0 ⎤
(C) (D) (B) x = ⎢ ⎥ x+ ⎢ ⎥u
( s − 1)( s + 1) ( s + 1) ⎣ −7 −2⎦ ⎣1 ⎦
Y = [3 1]x
6. The state space model for an electrical network is
shown in the figure where the current I and voltage ⎡1 0⎤ ⎡1 ⎤
(C) x = ⎢ ⎥ x+ ⎢ ⎥u
across the capacitor Vc are the state variables. ⎣ −2 −7⎦ ⎣0 ⎦
Y = [1 3] x
103 Ω
(D) None of these
1
10. From question 9, the transfer function of the system is
s s+3
10 volt 1nF (A) (B)
Vc s + 2s + 7
2 s + 2s + 7
2
10 volt
3 2s + 3
(C) (D)
s + 2s + 7
2 s + 2s + 7
2
3.336 | Part III • Unit 3 • Control Systems

Direction for questions 11 and 12: 13. A continuous time, linear time-invariant system is
The circuit diagram is shown in the figure. d2 y dy dx
described by +4 + 3y(t) = 2 + 4x
–
2F
+
dt dt dt
Assuming zero initial conditions, the response y(t)
of the above system for the input x(t) = e-2t u(t) is
5A ↑ 4Ω ±
3H 10 V given by
(A) (e-t – e-3t) u(t) (B) (e-t + e-3t) u(t)
t +3t
(C) (e - e ) u(t) (D) (et + e3t) u(t)
11. The state equations of the above circuit are 14. The system with the state equation
⎡. ⎤ ⎡ −1 0 ⎤ ⎡1⎤
iL ⎡ 0 −1/3⎤ ⎡ iL ⎤ ⎡1/3 0 ⎤ ⎡10 ⎤
(A) ⎢ ⎥ = ⎢ + x = Ax + Bu and A = ⎢ ⎥ ;B= ⎢ ⎥
⎢ . ⎥ ⎣ −1/2 −18/ ⎥⎦ ⎢⎣ vc ⎥⎦ ⎢⎣18
/ −1/2⎥⎦ ⎢⎣ 5 ⎥⎦ ⎣ 1 −2⎦ ⎣1⎦
⎢⎣ v c ⎥⎦
(A) System is controllable
⎡. ⎤ (B) System is uncontrollable
iL ⎡ 0 −1/3⎤ ⎡ iL ⎤ ⎡1/3 0 ⎤ ⎡10 ⎤
(B) ⎢ ⎥ = ⎢ + (C) System is controllable and observable
⎢ . ⎥ ⎣1/2 −18 / ⎥⎦ ⎢⎣ vc ⎥⎦ ⎢⎣18
/ −1/2⎥⎦ ⎢⎣ 5 ⎥⎦ (D) None of these
⎢⎣ v c ⎥⎦
15. Consider the state transition matrix:
⎡ iL ⎤ ⎡ −1/3 0 ⎤ ⎡ iL ⎤ ⎡ 0 1/3 ⎤ ⎡10 ⎤ ⎡ s +1 −1 ⎤
(C) ⎢ . ⎥ = ⎢ ⎢ v ⎥ + ⎢18 f(t) = L-1 [f(s)] = L-1 ⎢
⎢⎣ v c ⎥⎦ ⎣ 18
/ −1/2⎥⎦ ⎣ c ⎦ ⎣ / −1/2⎦
⎥ ⎢5⎥
⎣ ⎦ s + 3s + 2 s + 3s + 2
2 2 ⎥
⎢ ⎥
⎢ s+2 ⎥
⎢⎣ 0
(D) None of these s 2 + 3s + 2 ⎥⎦
12. From question 11, the eigen values are The eigen values of the system are
(A) –0.06 ± j0.06 (B) –0.06 ± j0.403 (A) 0 and – 2 (B) –1 and 2
(C) –0.403 ± j0.06 (D) –0.12 ± j0.12 (C) 1 and 2 (D) –1 and – 2

Previous Years’ Questions

⎡ −2 2 ⎤ (C) neither controllable nor observable
1. If A = ⎢ ⎥ , then sin At is[2004]
⎣ 1 −3⎦ (D) controllable and observable

1 ⎡ sin( −4t ) + 2 sin( −t ) −2 sin( −4t ) + 2 sin( −t ) ⎤ ⎡1 0 ⎤

(A) 3. Given A = ⎢ At
⎥ , the state transition matrix e is
3 ⎢⎣ − sin( −4t ) + sin( −t ) 2 sin( −4t ) + sin( −t ) ⎥⎦ ⎣0 1⎦
given by [2004]
⎡sin( −2t ) sin( 2t ) ⎤
(B) ⎢ ⎥ ⎡ 0 e −t ⎤ ⎡et 0⎤
⎣ sin(t ) sin( −3t ) ⎦ (A) ⎢
−t ⎥ (B) ⎢ ⎥
⎣e 0 ⎦ ⎣0 et ⎦
1 ⎡ sin( 4t ) + 2 sin(t ) 2 sin( −4t ) + 2 sin( −t ) ⎤
(C)
3 ⎢⎣ − sin( −4t ) + sin(t ) 2 sin( 4t ) + sin(t ) ⎥⎦ (C) ⎢
⎡e −t 0 ⎤
(D)
⎡ 0 et ⎤
−t ⎥ ⎢ t ⎥
⎣ 0 e ⎦ ⎣e 0 ⎦
(D) 1 ⎡ cos(t )2 cos(t ) −2 cos( −4t ) + 2 sin( −t ) ⎤
3 ⎣ − cos( −4t ) + sin( −t ) −2 cos( −44t ) + cos( −t ) ⎥⎦
⎢ 4. A linear system is described by the following state
equation:
2. The state variable equations of system are ⎡ 0 1⎤
• X (t ) = AX (t ) + BU (t ), A = ⎢ ⎥
⇒ X1 = –3x1 –x2 + u ⎣ −1 0 ⎦
• The state -transition matrix of the system is [2006]
⇒ X2 = 2x1 ⎡ cos t sin t ⎤ ⎡ − cos t sin t ⎤
(A) ⎢ (B) ⎢
y = x1 + u
⎣ − sin t cos t ⎥⎦ ⎣ − sin t − cos t ⎥⎦
The system is [2004]
(A) controllable but not observable ⎡ − cos t − sin t ⎤ ⎡cos t − sin t ⎤
(C) ⎢ (D) ⎢
(B) observable but not controllable ⎣ − sin t cos t ⎥⎦ ⎣cos t sin t ⎥⎦
 Chapter 5 • State Space Analysis | 3.337

5. The state space representation of a separately excited 7. The state variable representation of the system x can
DC servo motor dynamics is given as follows: be [2010]
⎡ dw ⎤ ⎡ 1 1⎤ ⎡0 ⎤
⎢ dt ⎥ ⎡ −1 1 ⎤ ⎡w ⎤ ⎡0 ⎤ X =⎢ ⎥ x + ⎢ ⎥u
⎢ ⎥ =⎢ ⎥ ⎢ ⎥ + ⎢ ⎥U (A) ⎣ −1 0 ⎦ ⎣ 2⎦
⎢ dia ⎥ ⎣ −1 − 10 ⎦ ⎣ia ⎦ ⎣10 ⎦
⎢⎣ dt ⎥⎦ y = [0 0.5] x

where w is the speed of the motor, ia is the armature ⎡ −1 1 ⎤ ⎡0 ⎤

current and U is the armature voltage. The transfer X =⎢ ⎥ x + ⎢ ⎥u
(B) ⎣ −1 0 ⎦ ⎣ 2⎦
function w( s) of the motor is [2007]
U ( s) y = [0 0.5] x

10 1
(A) (B) ⎡ 1 1⎤ ⎡0 ⎤
+ 11s + 11
s2 + 11s + 11
s2 X =⎢ ⎥ x + ⎢ 2⎥ u
(C) ⎣ −1 0 ⎦ ⎣ ⎦
10 s + 10 1
(C) (D) y = [0.5 0.5] x
s 2 + 11s + 11 s2 + s + 1
6. A single flow graph of a system is given as follows: ⎡ −1 1 ⎤ ⎡0 ⎤
−α X =⎢ ⎥ x + ⎢ ⎥u
(D) ⎣ −1 0 ⎦ ⎣ 2⎦
1 1/s y = [0.5 0.5] x
α
−β β
1/s
8. The transfer function of the system is [2010]
γ
1 s +1 s −1
1/s (A) (B)
s2 + 1 s2 + 1
−γ
s +1 s −1
The set of equations that correspond to this signal (C) (D)
s2 + s + 1 s2 + s + 1
flow graph is [2008]
⎛ x1 ⎞ ⎡ b −g 0 ⎤ ⎛ x1 ⎞ ⎡1 0 ⎤ 9. The block diagram of a system with one input u and
d ⎜ ⎟ ⎢ ⎛ u1 ⎞
(A) x2 = g a 0 ⎥⎥ ⎜ x2 ⎟ + ⎢⎢0 0 ⎥⎥ ⎜ ⎟ two outputs y1 and y2 is given below.
dt ⎜ ⎟ ⎢ ⎜ ⎟ ⎝ u2 ⎠
⎝ x3 ⎠ ⎢⎣ −a −b 0 ⎥⎦ ⎝ x3 ⎠ ⎢⎣0 1 ⎥⎦ 1
u y1
⎛ x1 ⎞ ⎡0 a g ⎤ ⎛ x1 ⎞ ⎡0 0 ⎤ s +2
d ⎜ ⎟ ⎢ ⎛ u1 ⎞
(B) x2 = 0 −a −g ⎥⎥ ⎜ x2 ⎟ + ⎢⎢0 1 ⎥⎥ ⎜ ⎟
dt ⎜ ⎟ ⎢ ⎜ ⎟ ⎝ u2 ⎠
⎝ x3 ⎠ ⎢⎣0 b − b ⎥⎦ ⎝ x3 ⎠ ⎢⎣1 0 ⎥⎦ 2
y2
s +2
⎛ x1 ⎞ ⎡ −a −b 0 ⎤ ⎛ x1 ⎞ ⎡1 0 ⎤
d ⎜ ⎟ ⎢ ⎛ u1 ⎞
(C) x2 = − b −g 0 ⎥⎥ ⎜ x2 ⎟ + ⎢⎢0 1 ⎥⎥ ⎜ ⎟ A state space model of the above system in terms of
dt ⎜ ⎟ ⎢ ⎜ ⎟ ⎝ u2 ⎠
⎝ x3 ⎠ ⎢⎣ a g 0 ⎥⎦ ⎝ x3 ⎠ ⎢⎣0 0 ⎥⎦ the state vector x and the output vector y =[y1 y2]T is
 [2011]
⎛ x1 ⎞ ⎡ −g 0 b ⎤ ⎛ x1 ⎞ ⎡0 1 ⎤ •
d ⎜ ⎟ ⎢ ⎛ u1 ⎞
(D) x2 = g 0 a ⎥⎥ ⎜ x2 ⎟ + ⎢⎢0 0 ⎥⎥ ⎜ ⎟ (A) X = [2]x + [1]u; y = [1 2] x
dt ⎜ ⎟ ⎢ ⎜ ⎟ ⎝ u2 ⎠
⎝ x3 ⎠ ⎢⎣ − b 0 −a ⎥⎦ ⎝ x3 ⎠ ⎢⎣1 0 ⎥⎦ •
(B) X = [-2]x + [1]u; y = ⎢⎡ ⎤⎥ x
1
Direction for questions 7 and 8: ⎣ 2⎦
•
−2 0 ⎤
(C) X = ⎡⎢
The signal flow graph of a system is shown below. ⎡1⎤
1 0 − 2 ⎥ x + ⎢ ⎥ u; y = [1 2]
⎣ ⎦ ⎣1⎦
2 1/s 0.5
1/s •
(D) X = ⎡⎢
U(s) Y(s) 2 0⎤ ⎡1⎤ u; y = ⎡1 ⎤
⎥x+ ⎢1⎥ ⎢ 2⎥ x
1 ⎣0 2⎦ ⎣⎦ ⎣ ⎦
−1

−1
3.338 | Part III • Unit 3 • Control Systems

Direction for questions 10 and 11: 13. An unforced linear time-invariant system is repre-
The state diagram of a system is shown below, A system sented by
is described by the state-variable equations ⎡• ⎤
⎢ x1 ⎥ = ⎡ −1 0 ⎤ ⎡ x1 ⎤
x = AX + Bu, y = CX + Du ⎢ • ⎥ ⎢⎣ 0 −2⎥⎦ ⎢⎣ x2 ⎥⎦
1 −1 1 −1 1 ⎢⎣ x 2 ⎥⎦
u y
If the initial conditions are x1(0) = 1 and x2(0) = -1,
1 1
the solution of the state equation is [2014]
s s (A) x1(t) = −1, x2(t) = 2
10. The state-variable equations of the system shown in (B) x1(t) = −e-t, x2(t) = 2e-t
the figure above are [2013] (C) x1(t) = e-t, x2(t) = -e-2t
(D) x1(t) = −e-t, x2(t) = -2e-t
• ⎡ −1 0 ⎤ ⎡ −1⎤
(A) X = ⎢ ⎥ X + ⎢ ⎥ u ; Y = [1 –1]X + U 14. Consider the state space system expressed by the sig-
⎣ 1 −1⎦ ⎣1⎦ nal flow diagram shown in the figure. [2014]
• ⎡ −1 0 ⎤ ⎡ −1⎤ c3 c2
(B) X = ⎢ ⎥ X + ⎢ 1 ⎥ u ; Y = [–1 –1]X + U
⎣ −1 −1⎦ ⎣ ⎦ −1
I s s−1 s−1
u x3 x2 x1 c1 y
• ⎡ −1 0 ⎤ ⎡ −1⎤ a3
a2
X =⎢ ⎥ X +⎢ ⎥u a1
(C) ⎣ −1 −1⎦ ⎣1⎦
y = [ −1 −1] X − u
The corresponding system is
• ⎡ −1 −1⎤ ⎡ −1⎤ (A) always controllable (B) always observable
X =⎢ ⎥ X +⎢ 1 ⎥u
(D) ⎣ 0 −1⎦ ⎣ ⎦ (C) always stable (D) always unstable
y = [1 −1] X − u 15. The state equation of a second-order linear system is
given by
11. The state transition matrix eAt of the system shown •
figure above is [2013] x(t ) = Ax(t ), x(0) = x0

⎡ e−t 0 ⎤ ⎡ e−t 0 ⎤ ⎡1⎤ ⎡ e−t ⎤

(A) ⎢ −1 − t ⎥ (B) ⎢ − t ⎥ For x0 = ⎢ ⎥ , x(t ) = ⎢ − t ⎥
⎣te e ⎦ ⎣ − te e−t ⎦ ⎣ −1⎦ ⎣ −e ⎦
⎡0⎤ ⎡ e − t − e −2t ⎤
⎡e − t 0 ⎤ ⎡e − t − te −1 ⎤ and for x0 = ⎢ ⎥ , x(t ) = ⎢ − t
(C) ⎢ − t ⎥ (D) ⎢ ⎥ −2t ⎥
e−t ⎦ e−t ⎦ ⎣1⎦ ⎣ − e + 2e ⎦
⎣e ⎣ 0
⎡3⎤
12. Consider the state space model of a system as given when x0 = ⎢ ⎥ , x (t ) is [2014]
below. [2014] ⎣ 5⎦

⎡ •⎤ ⎡ −8e − t + 11e −2t ⎤ ⎡ 11e − t − 8e −2t ⎤

(A) ⎢ − t −2t ⎥
(B) ⎢ −2t ⎥
⎢ x1 ⎥ ⎡ −1 1 0 ⎤ ⎡ x ⎤ ⎡ 0 ⎤ ⎣ 8e − 22e ⎦
−t
⎣ −11e + 16e ⎦
⎢• ⎥ ⎢ 1
⎢ x2 ⎥ = ⎢ 0 −1 0 ⎥⎥ ⎢⎢ x2 ⎥⎥ + ⎢⎢ 4 ⎥⎥ u ⎡ 3e − t − 5e −2t ⎤ ⎡ 5e − t − 3e −2t ⎤
⎢ ⎥ ⎢ (C) ⎢ ⎥ (D) ⎢ −2t ⎥
⎢ x• ⎥ ⎣ 0 0 −2⎥⎦ ⎢⎣ x3 ⎥⎦ ⎢⎣ 0 ⎥⎦
− − −t
⎣ −3e + 10e ⎦ ⎣ −5e + 6e ⎦
t 2t

⎢⎣ 3 ⎥⎦ 16. The state transition matrix Φ(t) of a system

⎡ x1 ⎤ ⎡ ⎤
⎢ x1 ⎥ = ⎡0 1 ⎤ ⎡ x1 ⎤ is
y = [1 1 1] ⎢⎢ x2 ⎥⎥ ⎢ ⎥ ⎢⎣0 0 ⎥⎦ ⎢⎣ x2 ⎥⎦
[2014]
⎢⎣ x3 ⎥⎦ ⎣⎢ x 2 ⎦⎥
The system is
⎡t 1 ⎤ ⎡1 0 ⎤
(A) controllable and observable (A) ⎢ ⎥ (B) ⎢ ⎥
(B) uncontrollable and observable ⎣1 0 ⎦ ⎣t 1 ⎦
(C) uncontrollable and unobservable ⎡0 1⎤ ⎡1 t ⎤
(D) controllable and unobservable (C) ⎢ ⎥ (D) ⎢ ⎥
⎣1 t ⎦ ⎣0 1⎦
 Chapter 5 • State Space Analysis | 3.339

Direction for questions 17 and 18: 19. The network is described by the model is
Consider a linear system whose state space representation •
•
is X (t) = AX(t). If the initial state vector of the system is x1 = 2x1 − x2 + 3u
•
⎡1⎤ ⎡ e −2t ⎤ x 2 = −4x2 − u
X(0) = ⎢ ⎥ , then the system response is X(t) = ⎢ −2t ⎥
.
⎣ −2⎦ ⎣ −2e ⎦ y = 3x1 − 2x2
If the initial state vector of the system changes to X(0) =
⎛ Y ( s) ⎞
⎡1⎤ ⎡ e −t ⎤ The transfer function H(s) ⎜ = is [2015]
⎢ −1⎥ , then the system response becomes X(t) = ⎢ − t ⎥ ⎝ U ( s) ⎟⎠
⎣ ⎦ ⎣ −e ⎦
11s + 35 11s − 35
17. The eigen value and eigen vector pairs (li, ui) for the (A) (B)
( s − 2)( s + 4) ( s − 2)( s + 4)
system are [2014]
⎛ ⎡ 1 ⎤⎞ ⎛ ⎡ 1 ⎤⎞ 11s + 38 11s − 38
(C) (D)
(A) ⎜ −1, ⎢ ⎥⎟ and ⎜ −2, ⎢ ⎥⎟ ( s − 2)( s + 4) ( s − 2)( s + 4)
⎝ ⎣ −1⎦⎠ ⎝ ⎣ −2⎦⎠
⎛ ⎡ 1 ⎤⎞ ⎛ ⎡ 1 ⎤⎞ 20. A second order linear time invariant system is
(B) ⎜ −1, ⎢ ⎥⎟ and ⎜⎝ 2, ⎢ −2⎥⎟⎠ described by the following state equations
⎝ ⎣ −1⎦⎠ ⎣ ⎦
d
⎛ ⎡ 1 ⎤⎞ ⎛ ⎡ 1 ⎤⎞ x (t) + 2x1(t) = 3u(t)
(C) ⎜ −2, ⎢ ⎥⎟ and ⎜ −2, ⎢ ⎥⎟ dt 1
⎝ ⎣ −1⎦⎠ ⎝ ⎣ −2⎦⎠ d
⎛ x (t) + x2(t) = u(t)
⎡ 1 ⎤⎞ ⎛ ⎡ 1 ⎤⎞ dt 2
(D) ⎜ −2, ⎢ ⎥⎟ and ⎜⎝1, ⎢ −2⎥⎟⎠
⎝ ⎣ −1⎦⎠ ⎣ ⎦ Where x1(t) and x2(t) are the two state variables and
18. The system matrix A is [2014] u(t) denotes the input. If the output c(t) = x1(t), then
the system is [2016]
⎡ 0 1⎤ ⎡1 1⎤ (A) Controllable but not observable
(A) ⎢ ⎥ (B) ⎢ ⎥
⎣ −1 1⎦ ⎣ −1 −2⎦ (B) Observable but not controllable
(C) Both controllable and observable
⎡2 1⎤ ⎡0 1⎤ (D) Neither controllable nor observable
(C) ⎢ ⎥ (D) ⎢ ⎥
⎣ −1 −1⎦ ⎣ −2 −3⎦

Answer Keys
Exercises
Practice Problems 1
1. D 2. C 3. A 4. B 5. C 6. D 7. B 8. C 9. A 10. C
11. A 12. C 13. B 14. B 15. B

Practice Problems 2
1. B 2. A 3. D 4. C 5. D 6. C 7. A 8. C 9. B 10. B
11. B 12. B 13. A 14. B 15. D

Previous Years’ Questions

1. A 2. D 3. B 4. A 5. A 6. D 7. D 8. C 9. B 10. A
11. A 12. B 13. C 14. A 15. B 16. D 17. A 18. D 19. A 20. A
3.340 | Part III • Unit 3 • Control Systems

Test

Control Systems Time: 60 Minutes

Direction for questions 1 to 30: Select the correct alterna- (A) −6. e-3t + e−t (B) e−t + 2/3 e−3t
tive from the given choices. (C) 6. e − e
3t −t (D) None of these
1. The transfer function gain between C(S) and R(S) in the 10. The impulse response of an initially relaxed system is
following figure is e-4t u(t). To produce a response of t .e-4t .u(t), the input
–4 must be equal to
2 3 (A) e +4t u(−t) (B) e−4t u(t)
(C) t.e −4t (D) e−4t. u(t)
R(s) C(s)
2 11. The closed loop gain of the system shown in the fol-
lowing figure is
(A) 5 (B) 2.5 (C) 10 (D) 2
2. The Laplace transform of a transportation lag of 8 s is 2

(A) exp (8s) (B) exp (–8s) R(s) + C(s)

1 × 4
(C) (D) exp (–s/8) –
3−8
3. The transfer function of ZOH (zero-order hold) is 1/2
(A) 1 – e-Ts (B) 1 – eTs
1− eTs 1− e −Ts (A) 2 (B) 3 (C) 6/4 (D) −4/6
(C) (D)
s s 12. The response c(t) of a system to an input r(t) is given by
4. The main drawback of a feedback system is the following differential equation
(A) inaccuracy (B) inefficiency d 2 c (t ) dc (t )
(C) unstability (D) insensitivity + c (t ) = 2.r (t )
+ 3.
dt 2 dt
5. The transfer function of linear control system is defined The transfer function of the system is given by
as the
(A) Fourier transform of impulse response (A) 3 (B) −2
(B) Laplace transform of unit step response s2 + 2s + 1 s2 + 3s + 2
(C) Laplace transform of impulse response
(D) None of these (C) 2 2 (D)
6. Transfer function is defined for s 2 + 3s + 3 s 2 + 3s + 1
(A) linear and time-variant system 13. Given open loop transfer function G(s) = 1 − s . The
(B) linear and time-invariant system s ( s + 1)
(C) non-linear and time-variant system system with the transfer function is operated in a closed
(D) non-linear and time-invariant system loop with unity feedback. The closed loop system is
7. The minimum phase transfer function is one having (A) unstable (B) stable
(A) poles and zeros in RHS of s-plane (C) marginally stable (D) conditionally stable
(B) poles and zeros in LHS s-plane 14. The signal flow graph of the system is shown in the
(C) poles in LHS and zeros in RHS s-plane
given figure. The transfer function ( ) of the system
C s
(D) poles in RHS and zeros in RHS s-plane
D (s)
8. The unit impulse response of a unit feedback control is
system is given by c(t) = –e-t + 3e+t (t ≥ 0). The transfer
function is equal to D(s)

(A) ( s + 2) (B) s − 1
2
–1
s2 −1 s( s + 1)

(C) 2(s - 2) (D) 2( s + 2)

R(s) 1 2 s 4 C(s)
s2 + 1 s2 −1
9. The unit step response of a unit feedback control sys-
tem is given by c(t) = −e-t + 2. e−3t (t ≥ 0). The impulse
response is –3
 Test | 3.341

(A)
8s
(B)
4s C (s)
The transfer function is
1 + 6s 1 + 6s R (s)
−4 s −4 s
(C) (D) G1 G2
1 − 6s 1 + 6s (A)
1+ G1 H1 + G2 H 2
1− s
15. If a system has an open-loop transfer function ,
1+ s G1 G2
(B)
then the gain of the system at frequency of 1 rad/s will 1+ G1G2 + G2 H 2 + G1 H1 + G1G2 H1 H 2
be
(A) 1 (B) 0 (C) –1 (D) 1/2 (C) G1G2 + 1
1 + G1 H1 + G2 H 2 + G1G2 H1 H 2
16. The following block diagram is equivalent to
(D) None of these
C(s)
+ G
X1 ± 19.
G3
X2

(A) C(s) R(s) G1 1 G2 1 C(s)

G +
X1 ±

1/G X2 –H2

–H1
(B) X1 C(s)
G +
± The number of forward paths and individual loops are,
respectively,
X2
(A) 2, 2 (B) 2, 3 (C) 3, 2 (D) 2, 4
(C) X1 C(s) 20. 2
G +
±
5 + + + C(s)
G X2 3/s 4/s
–
–
(D) B and C
17. For what value of K, are the two block diagrams, as +
5 6s
shown in the following figure, equivalent? +

R(s) The output C(s) is

+ C(s)
4 ( 2 s + 15) (2s + 15)
1
K
S+2 + (A) (B)
5s (5s + 52) s (5s + 52)
4 ( 2 s − 15)
(C) (D) None of these
5s ( s + 52)
Figure 1
21. The closed-loop transfer function of a control system is
R(s)
s+3 C(s) C (s) 2 ( s + 1)
given by = for a unit step input the
s+2 R (s) ( s + 2) ( s − 3)
Figure 2 output is
(A) 1 (B) –1 (C) 3 + 1 (D) 3 + 2 ⎡ 1 1 8 ⎤
(A) ⎢ − + e −2t + e3t ⎥ u(t )
18. ⎣ 3 5 15 ⎦
(B) 0
R(s) + – G1 G2 (C) ⎣⎡ −1 + e −2t + e3t ⎤⎦ u(t )
C(s)
– –
⎡ 1 1 8 ⎤
(D) ⎢ − + e −2t + e −3t ⎥ u(t )
H1 H2 ⎣ 5 3 15 ⎦
3.342 | Part III • Unit 3 • Control Systems

22. –1 4 Direction for questions 27 and 28:

For the SFG shown in the following figure:
R(s) 1 3 1 5 C(s)
–H1

–2 G2 2
G5
R(s) 1 C(s)
1
2
C (s) 1
G1
3 G4
In the signal flow graph, the ratio between is
equal to R (s)
(A) –15 (B) –10 (C) 20 (D) –25
23. A system has a single pole at origin. Its impulse –H2
response will be
27. The number of forward path gains are
(A) constant (B) ramp
(A) 3 (B) 4 (C) 5 (D) 6
(C) decaying exponential (D) oscillating
28. The number of individual loop gains are
1
24. An open loop system has a transfer function . (A) 1 (B) 2
s ( s − 2) (C) 3 (D) None of these
It is converted into a closed loop system by providing
a negative feedback having transfer function (2s + 1).
The open loop and closed loop systems are, respectively, Direction for questions 29 and 30:
(A) stable and unstable
(B) unstable and stable –2
(C) unstable and unstable
(D) unstable and marginally stable
3 1
25. The unit ramp response of a system is 1 − e−t (1 + t).
Which is this system? R(s) 1 4 C(s)
5 2/s 3K
(A) unstable (B) stable –
(C) marginally stable (D) None of these
Y (s) s –1
26. The transfer function of the linear time-invariant
X (s)
system shown in the following figure C (s)
29. The transfer function is
+ R (s)
× G1(s)
+
X(s) 378 K 360 K
(A) (B)
15s 2 + 3s + 6 5s 2 + 12 s + 360 K
+
+ 378 K
× G2 (s) Y(s) (C) (D) None of these
+ 18s + 6
5s 2
30. The sensitivity of the transfer function with parameter
G1 ( s ) [G2 ( s ) + 1] G2 ( s ) [G1 ( s ) + 1]
(A) (B) ‘K’ is
1 − G1 ( s ) .G2 ( s ) 1 − G1 ( s ) .G2 ( s )
378 −378
(A) (B)
G1 ( s ) [G2 ( s ) + 1] G2 ( s ) [G1 ( s ) + 1] ( )
2
(C) (D) 5s 2 + 13s + 6 s2 + 13s + 6
1 − G1 ( s ) + G2 ( s ) 1 + G1 ( s ) .G2 ( s )
(C) 1 (D) –1

Answer Keys
1. D 2. B 3. D 4. C 5. C 6. B 7. B 8. D 9. A 10. B
11. C 12. D 13. C 14. D 15. A 16. C 17. A 18. B 19. A 20. A
21. A 22. D 23. A 24. D 25. B 26. B 27. C 28. A 29. A 30. C