Control Engineering

(제어공학)
Spring 2026
Controls 4

1
Chap. 3 State Variable Models
3.1 Introduction
3.2 The State Variables of a Dynamic System
3.3 The State Differential Equation
3.4 Signal-Flow Graph and Block Diagram Models
3.5 Alternative Signal-Flow Graph and Block Diagram Models
3.6 The Transfer Function from the State Equation
3.7 The Time Response and the State Transition Matrix
3.8 Design Examples
3.9 Analysis of State Variable Models Using Control Design Software
3.10 Sequential Design Example: Disk Drive Read System
3.11 Summary
2
3.1 Introduction
• A time-varying control system is a system in which one or more of
the parameters of the system may vary as a function of time.

• The time domain is the mathematical domain that incorporates the

response and description of a system in terms of time, t.

3
3.2 The State Variables of a Dynamic System
• The state of a system is a set of variables whose values, together with
the input signals and the equations describing the dynamics, will
provide the future state and output of the system.

• The state 𝐱(𝑡0 ) of a system at time 𝑡0 is the information at 𝑡0 that,

together with the input 𝐮(𝑡), for 𝑡 ≥ 𝑡0 , determines uniquely the
output 𝐲(𝑡) for all 𝑡 ≥ 𝑡0

4
3.2 The State Variables of a Dynamic System

System block diagram

5
3.2 The State Variables of a Dynamic System
• The state variables describe the present configuration of a system
and can be used to determine the future response, given the
excitation inputs and the equations describing the dynamics.

• The state summarizes the effect of past input on future output: Using
the state at 𝑡0 , we can express the input and output of a system as

State-input-output pairs
6
3.2 The State Variables of a Dynamic System

Dynamic system

7
3.2 The State Variables of a Dynamic System

A spring-mass-damper system 8
3.2 The State Variables of a Dynamic System

An RLC circuit

9
3.2 The State Variables of a Dynamic System
• Example: State of the circuit at time t0 → initial state
𝑥1 (𝑡0 )
𝐱(𝑡0 ) = 𝑥2 (𝑡0 )
𝑥3 (𝑡0 )

x1, x2, x3: state variables

Circuit with three state variables. 10

3.3 The State Differential Equation
• First-order differential equations can be written in general form as:

11
3.3 The State Differential Equation
Spring mass function Rewrite

Let’s give a name

organize

Final form

12
3.3 The State Differential Equation
• The system can be represented by the compact notation of the state
differential equation as
𝐱ሶ 𝑡 = 𝐀𝐱 𝑡 + 𝐁𝐮 𝑡

• Output equation:

𝑦 𝑡 = 𝐂𝐱 𝑡 + 𝐃𝐮(𝑡)

13
3.3 The State Differential Equation
• The matrix exponential function is defined as:

14
3.4 Signal-flow Graph and Block Diagram Models

RLC network (Signal-flow graph)

RLC network (Block diagram) 15

3.4 Signal-flow Graph and Block Diagram Models

Flow graph nodes and integrators for fourth-order system

16
3.4 Signal-flow Graph and Block Diagram Models
• Consider the fourth-order transfer function

• Define the four state variables as follows:

17
3.4 Signal-flow Graph and Block Diagram Models

FIGURE 3.9 Model for G(s) of Equation (3.45). (a) Signal-flow graph. (b) Block
18
diagram.
3.4 Signal-flow Graph and Block Diagram Models
• Phase variable canonical form

FIGURE 3.10 Model for G(s) of Equation (3.46) in the phase variable format. (a)
19
Signal-flow graph. (b) Block diagram
3.4 Signal-flow Graph and Block Diagram Models
• Phase variable canonical form

Fourth order Transfer function

FIGURE 3.10 Model for G(s) of Equation (3.46) in the phase variable format. (a)
20
Signal-flow graph. (b) Block diagram
3.4 Signal-flow Graph and Block Diagram Models

FIGURE 3.11 Flow graph of Figure 3.10 with nodes inserted.

22
3.4 Signal-flow Graph and Block Diagram Models
• In matrix form:

23
3.4 Signal-flow Graph and Block Diagram Models
• Alternative form: Input feedforward canonical form

FIGURE 3.12 (a) Alternative flow graph state model for Equation (3.46). This model
is called the input feedforward canonical form. (b) Block diagram of the input
24
feedforward canonical form.
3.4 Signal-flow Graph and Block Diagram Models
• Input feedforward canonical form:

• Matrix form (dual form): cf: phase variable form

26
3.4 Signal-flow Graph and Block Diagram Models
• Example 3.2 Two state variable models

FIGURE 3.13 (a) Phase variable flow graph state model for T(s). (b) Block diagram
27
for the phase variable canonical form.
3.4 Signal-flow Graph and Block Diagram Models

FIGURE 3.14 (a) Alternative flow graph state model for T(s) using the input
feedforward canonical form. (b) Block diagram model.
28
3.5 Alternative Signal-Flow Graph and Block Diagram Models

FIGURE 3.15 A block diagram model of an open-loop DC motor control with velocity
as the output.

29
3.5 Alternative Signal-Flow Graph and Block Diagram Models

FIGURE 3.16 (a) The physical state variable signal-flow graph for the block diagram
of Figure 3.15. (b) Physical state block diagram.
30
3.5 Alternative Signal-Flow Graph and Block Diagram Models

• The decoupled state variable model

where

• Diagonal canonical form for distinct poles

• Jordan canonical (block diagonal) form for
repeated poles

31
 3.5 Alternative Signal-Flow Graph and Block Diagram Models

FIGURE 3.17 (a) The decoupled state FIGURE 3.17 (b) The decoupled
variable flow graph model for the system state variable block diagram model
shown in block diagram form in Figure 3.15 32
 3.5 Alternative Signal-Flow Graph and Block Diagram Models
𝑙 sin𝜃 ≈ 𝑙𝜃
• Example 3.3 Inverted pendulum control

FIGURE 3.18 An inverted pendulum balanced FIGURE 3.19 A cart and an inverted
on a person’s hand by moving the hand to pendulum. The pendulum is constrained to
reduce θ(t). Assume, for ease, that the pivot in the vertical plane.
33
pendulum rotates in the x–y plane.
3.5 Alternative Signal-Flow Graph and Block Diagram Models
• Assume that M >> m and the angle of rotation 𝜃(𝑡), is small so that
the equations are linear.
• The sum of the forces in the horizontal direction is
′′
𝑀𝑦ሷ = 𝑢 − 𝑚 𝑙sin𝜃 →
• The sum of the torques about the pivot point is
𝑚𝑙 2 𝜃 ′′ = 𝑚𝑔𝑙sin𝜃 − 𝑚𝑦 ′′ 𝑙cos𝜃 →
• Let
• Then
Solve for x2 with by substituting lx4

34
3.5 Alternative Signal-Flow Graph and Block Diagram Models
• Assume that M >> m and the angle of rotation 𝜃(𝑡), is small so that
the equations are linear.
• The sum of the forces in the horizontal direction is
′′
𝑀𝑦ሷ = 𝑢 − 𝑚 𝑙sin𝜃 →
• The sum of the torques about the pivot point is
𝑚𝑙 2 𝜃 ′′ = 𝑚𝑔𝑙sin𝜃 − 𝑚𝑦 ′′ 𝑙cos𝜃 →
• Let
• Then
Solve for x2 with by substituting lx4

Solve by substituting x 2
35
3.6 The Transfer Function from The State Equation
𝐱ሶ 𝑡 = 𝐀𝐱(𝑡) + 𝐁𝑢(𝑡) L.T. (𝑠𝐈 − 𝐀)𝐗 𝑠 = 𝐁𝑈(𝑠)
𝑦(𝑡) = 𝐂𝐱(𝑡) + 𝐃𝑢(𝑡) 𝑌 𝑠 = 𝐂𝐗 𝑠 + 𝐃𝑈(𝑠)
−1
𝑠𝐈 − 𝐀 = 𝚽(𝑠)

𝐗 𝑠 = 𝚽(𝑠)𝐁𝑈(𝑠)
𝑌 𝑠 = [𝐂𝚽(𝑠)𝐁 + 𝐃]𝑈(𝑠)
• The transfer function G(s)
𝑌 𝑠 −1
𝐺 𝑠 = = 𝐂 𝑠𝐈 − 𝐀 𝐁 + 𝐃 = 𝐂𝚽(𝑠)𝐁 + 𝐃
𝑈 𝑠

36
3.6 The Transfer Function from The State Equation
• Example 3.4 Transfer function of an RLC circuit

37
3.6 The Transfer Function from The State Equation

where
• The transfer function is

39
3.7 The Time Response and The State Transition Matrix

−1
𝚽 𝑠 = 𝑠𝐈 − 𝐀
(𝐔 𝑠 = 0)

Solution for second-order system

40
3.7 The Time Response and The State Transition Matrix
• Example 3.5 Evaluation of the state transition matrix
• RLC network: R = 3, L = 1, and C = 1/2

where
41
 3.7 The Time Response and The State Transition Matrix

R = 3, L = 1, and C = 1/2

FIGURE 3.20 Flow graph of

the RLC network.

FIGURE 3.21 Flow graph of the RLC network with U(s) = 0.

42
3.7 The Time Response and The State Transition Matrix
• Mason’s signal-flow gain formula
σ𝑘 𝑃 𝑘 𝑠 Δ𝑘 𝑠
• 𝑇 𝑠 =
Δ(𝑠)
• From 𝑥1 0 → 𝑋1 (𝑠):

43
3.7 The Time Response and The State Transition Matrix
• Mason’s signal-flow gain formula
• 𝑥2 0 → 𝑋1 (𝑠):

• 𝑥1 0 → 𝑋2 (𝑠):

• 𝑥2 0 → 𝑋2 (𝑠):

44
3.7 The Time Response and The State Transition Matrix
• The state transition matrix

45
3.7 The Time Response and The State Transition Matrix

Time response of the state variables of the RLC

network for x1(0) = x2(0) = 1
46
3.7 The Time Response and The State Transition Matrix

Trajectory of the state vector in the (x1, x2)-plane

47
3.8 Design Examples
• Example 3.6 Modeling the orientation of a space station

Figure 3.24 The International Space Station

48
3.8 Design Examples

49
3.8 Design Examples
• Example 3.7 Printer belt drive modeling

FIGURE 3.26 Printer belt-drive system.

50
3.8 Design Examples
• Example 3.7 Printer belt drive modeling

FIGURE 3.27 Printer belt-drive model.

51
3.8 Design Examples

FIGURE 3.28 Printer belt drive. (a) Signal-flow graph. (b) Block diagram model.
52
3.8 Design Examples

FIGURE 3.28 Printer belt drive. (a) Signal-flow graph. (b) Block diagram model.
53
3.8 Design Examples
• Block diagram reduction

FIGURE 3.29 Printer belt drive block diagram reduction.

54
3.8 Design Examples

FIGURE 3.29 (continued) Printer belt drive block diagram reduction.

55
3.8 Design Examples

FIGURE 3.29 (continued) Printer belt drive block diagram reduction.

56
3.8 Design Examples

57
3.8 Design Examples
• With a step disturbance, Td(s) = a/s,

• For an average value of k = 20,

• Taking the inverse Laplace transform

58
3.8 Design Examples

Response of x1(t) to a step disturbance: peak value = -0.0325

• reduces the effect of the unwanted disturbance to a relatively

small magnitude → achieved design objective 59
3.9 Analysis of State Variable Models Using Control Design
Software
• The time-domain method utilizes a state-space representation of the
system model, given by

𝐱ሶ 𝑡 = 𝐀𝐱 𝑡 + 𝐁𝑢 𝑡

𝑦 𝑡 = 𝐂𝐱 𝑡 + 𝐃𝑢(𝑡)

60
3.9 Analysis of State Variable Models Using Control Design
Software

FIGURE 3.31 (a) The ss function. (b) Linear system model conversion.
61
3.9 Analysis of State Variable Models Using Control Design
Software

FIGURE 3.32 Conversion of Equation (3.115) to a state-space representation. (a)

m-file script. (b) Output printout. 62
3.9 Analysis of State Variable Models Using Control Design
Software

FIGURE 3.33 Block diagram with X1(s) defined as the leftmost state variable. 63
3.9 Analysis of State Variable Models Using Control Design
Software
RLC network

FIGURE 3.34 Computing the state transition matrix for a given time, ∆t = dt.

64
3.9 Analysis of State Variable Models Using Control Design
Software

FIGURE 3.35 The lsim function for calculating the output and state response. 65
3.9 Analysis of State Variable Models Using Control Design
Software

FIGURE 3.36 Computing the time response for nonzero initial conditions (1, 1)
and zero input using lsim. 66
3.9 Analysis of State Variable Models Using Control Design
Software

FIGURE 3.36 (continued) Computing the time response for nonzero initial
conditions (1, 1) and zero input using lsim.
67
3.10 Sequential Design Example: Disk Drive Read System

FIGURE 3.37 (a) Model of the two-mass system with a spring flexure. (b)
Simplified model with a rigid spring.
68
3.10 Sequential Design Example: Disk Drive Read System
TABLE 3.1 Typical parameters of the Two-Mass Model

69
3.10 Sequential Design Example: Disk Drive Read System

FIGURE 3.38 Transfer function model of head reader device with a rigid spring.

70
3.10 Sequential Design Example: Disk Drive Read System
• State variable model of the two-mass system

71
 3.10 Sequential Design Example: Disk Drive Read System

FIGURE 3.39 Response of 𝑦 ̇ for a step input for the two-mass model with k = 10.
72
 3.10 Sequential Design Example: Disk Drive Read System

FIGURE 3.39 (continued) Response of 𝑦 ̇ for a step input for the two-mass model
73
with k = 10.
3.11 Summary
• State variable feedback:
• The use of a control signal formed as a direct function of all the
state variables.
• State variable models:
• The system can be represented by the compact notation of the
state differential equation
• Canonical forms: Phase variable (≈ Controllable), Input
feedforward ( = Observable)
• Decoupled state variable model: Diagonal canonical, Jordan
canonical
• Transfer function from the state equation: state transition matrix
74