18/03/2024

System Dynamics and Control 1 Steady-State Error System Dynamics and Control 2 Steady-State Error

Learning Outcome
After completing this chapter, the student will be able to
• Find the steady-state error for a unity feedback system
• Specify a system’s steady-state error performance
• Design the gain of a closed-loop system to meet a steady-state
error specification
• Find the steady-state error for disturbance inputs
• Find the steady-state error for nonunity feedback systems
• Find the steady-state error sensitivity to parameter changes
• Find the steady-state error for systems represented in state space

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 3 Steady-State Error System Dynamics and Control 4 Steady-State Error

§1.Introduction §1.Introduction
- Control systems analysis and design focus on three specifications Definition and Test Inputs
• transient response - Steady-state error is the difference between the input and the
• stability output for a prescribed test input as 𝑡 → ∞. Test inputs used for
• steady-state errors steady-state error analysis and design are summarized
Test waveforms for evaluating steady-state errors of position control systems
taking into account the robustness of the design along with Waveform Name Physical interpretation Time function Laplace transform
𝑟(𝑡)
economic and social considerations
1
- Control system design entails trade-offs between desired step constant position 1
𝑡 𝑠

• transient response, 𝑟(𝑡)

1
• steady-state error, and 𝑡
ramp constant velocity 𝑡
𝑠2
• the requirement that the system be stable 𝑟(𝑡)
1 2 1
parabola constant acceleration 𝑡
𝑡 2 𝑠3

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 5 Steady-State Error System Dynamics and Control 6 Steady-State Error

§1.Introduction §1.Introduction
Test waveforms for evaluating steady-state errors of position control systems Test waveforms for evaluating steady-state errors of position control systems
Waveform Name Physical interpretation Time function Laplace transform Waveform Name Physical interpretation Time function Laplace transform
𝑟(𝑡) 𝑟(𝑡)
1 1
step constant position 1 ramp constant velocity 𝑡
𝑡 𝑠 𝑡 𝑠2

satellite in geostationary orbit • Step inputs represent constant position satellite in geostationary orbit • Ramp inputs represent constant-velocity
satellite orbiting at constant velocity satellite orbiting at constant velocity
and thus are useful in determining the inputs to a position control system, and
accelerating missile accelerating missile
ability of the control system to position can be used to test a system’s ability to
itself with respect to a stationary target, follow a linearly increasing input or,
tracking system tracking system
such as a satellite in geostationary orbit equivalently, to track a constant velocity
Ex.: An antenna position control is a target
system that can be tested for accuracy Ex.: A position control system that tracks
using step inputs a satellite that moves across the sky at a
constant angular velocity, would be tested
with a ramp input to evaluate the steady-
state error between the satellite’s angular
position and that of the control system
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

1
 18/03/2024

System Dynamics and Control 7 Steady-State Error System Dynamics and Control 8 Steady-State Error

§1.Introduction §1.Introduction
Test waveforms for evaluating steady-state errors of position control systems
Waveform Name Physical interpretation Time function Laplace transform Evaluating Steady-State Errors
𝑟(𝑡)
𝑐(𝑡) 𝑒2 (∞)
1 2 1 input
parabola constant acceleration 𝑡
𝑡 2 𝑠3 output 1
output 2
satellite in geostationary orbit • Parabola inputs represent constant
satellite orbiting at constant velocity
acceleration inputs to position control
accelerating missile
systems and can be used to represent
𝑡
accelerating targets, such as the Steady-state error with step input
tracking system
missile, to determine the steady-state
error performance
- Consider the system with step input
Application to Stable Systems • output 1 has zero steady-state error
Since we are concerned with the difference between the input • output 2 has a finite steady-state error, 𝑒2 (∞)
and the output of a feedback control system after the steady
state has been reached, our discussion is limited to stable
systems, where the natural response approaches zero as 𝑡 → ∞
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 9 Steady-State Error System Dynamics and Control 10 Steady-State Error

§1.Introduction §1.Introduction
- Consider the system with ramp input - Consider the closed-loop control system error as the following
𝑐(𝑡) 𝑒2 (∞) 𝑐(𝑡) 𝑒2 (∞) 𝑅(𝑠) 𝐶(𝑠) 𝐸(𝑠) 𝑅(𝑠) 𝐸(𝑠) 𝐶(𝑠)
input
𝑇(𝑠) 𝐺(𝑠)
output 1 input
output 2
output 1 Closed-loop control system error Closed-loop control system error
output 2 a. General representation b. Representation for unity feedback systems

output 3 𝐺(𝑠) : system transfer function

𝑡 𝑡
Steady-state error with step input Steady-state error with ramp input
𝑇(𝑠) : closed-loop transfer function
𝐸(𝑠) : error, the difference between the input and the output
• output 1 has zero steady-state error
→ study the steady-state, or final, value of 𝑒(𝑡)
• output 2 has a finite steady-state error, 𝑒2 (∞), as measured
vertically after the transients have died down • First, study and derive expressions for the steady-state error
for unity feedback systems
• output 3 has infinite steady-state error, as measured
vertically after the transients have died down, if the output’s • Then, expand to nonunity feedback systems
slope is different from that of the input
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 11 Steady-State Error System Dynamics and Control 12 Steady-State Error

§1.Introduction §2.Steady-State Error for Unity Feedback Systems

Sources of Steady-State Error Steady-State Error in Terms of 𝑇(𝑠)
- Many steady-state errors in control systems arise from 𝑅(𝑠) 𝐶(𝑠) 𝐸(𝑠)
nonlinear sources, such as backlash in gears or a motor that 𝑇(𝑠)
will not move unless the input voltage exceeds a threshold Closed-loop control system error – general representation
→ study the steady state errors that arise from the configuration The error between the input, 𝑅(𝑠), and the output, 𝐶(𝑠)
of the system itself and the type of applied input 𝐸 𝑠 = 𝑅 𝑠 − 𝐶(𝑠) (7.2)
• steady-state error for unity feedback systems 𝐶 𝑠 = 𝑅 𝑠 𝑇(𝑠) (7.3)
• static error constants and system type → 𝐸 𝑠 = 𝑅 𝑠 [1 − 𝑇(𝑠)] (7.4)
Applying the final value theorem
𝑒 ∞ = lim 𝑒(𝑡)
𝑡→∞
= lim 𝑠𝐸(𝑠) (7.5)
𝑠→0
= lim 𝑠𝑅 𝑠 [1 − 𝑇(𝑠)] (7.6)
𝑠→0

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

2
 18/03/2024

System Dynamics and Control 13 Steady-State Error System Dynamics and Control 14 Steady-State Error

§2.Steady-State Error for Unity Feedback Systems §2.Steady-State Error for Unity Feedback Systems
- Ex.7.1 Steady-State Error in Terms of 𝑇(𝑠) Steady-State Error in Terms of 𝐺(𝑠)
Find the steady-state error for the system 𝑇 𝑠 = 5/(𝑠2 + 7𝑠 + 10) 𝑅(𝑠) 𝐸(𝑠) 𝐶(𝑠)
𝐺(𝑠)
if the input is a unit step
𝑅(𝑠) 𝐶(𝑠) 𝐸(𝑠)
Solution 𝑇(𝑠) Closed-loop control system error –representation for unity feedback systems
From the problem statement Consider the unity feedback control system
5 1 𝐸 𝑠 = 𝑅 𝑠 − 𝐶(𝑠) (7.8)
𝑇 𝑠 = 2 , 𝑅 𝑠 =
𝑠 + 7𝑠 + 10 𝑠 𝐶 𝑠 = 𝐸 𝑠 𝐺(𝑠) (7.9)
𝑠 2 + 7𝑠 + 5 𝑅(𝑠)
The error 𝐸 𝑠 = 𝑅 𝑠 1 − 𝑇 𝑠 = →𝐸 𝑠 = (7.10)
𝑠(𝑠 2 + 7𝑠 + 10) 1 + 𝐺(𝑠)
Since 𝑇(𝑠) is stable and, subsequently, 𝐸(𝑠) does not have Apply the final value theorem
RHP poles or 𝑗𝜔 poles other than at the origin, apply the final 𝑠𝑅(𝑠)
value theorem 𝑒 ∞ = lim 𝑠𝐸(𝑠) = lim (7.11)
𝑠→0 𝑠→0 1 + 𝐺(𝑠)
𝑠 2 + 7𝑠 + 5 1
𝑒 ∞ = lim 𝑠𝐸(𝑠) = lim 2 =
𝑠→0 𝑠→0 𝑠 + 7𝑠 + 10 2
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 15 Steady-State Error System Dynamics and Control 16 Steady-State Error

§2.Steady-State Error for Unity Feedback Systems §2.Steady-State Error for Unity Feedback Systems
Step Input Ramp Input
𝑠(1/𝑠) 1 𝑠(1/𝑠 2) 1
𝑒 ∞ = 𝑒step ∞ = lim = (7.12) 𝑒 ∞ = 𝑒ramp ∞ = lim = (7.16)
𝑠→0 1 + 𝐺(𝑠) 1 + lim 𝐺(𝑠) 𝑠→0 1 + 𝐺(𝑠) lim 𝑠𝐺(𝑠)
𝑠→0 𝑠→0
In order to have zero steady-state error, the dc gain, lim 𝐺(𝑠), To have zero steady-state error for a ramp input
𝑠→0
of the forward transfer function must → ∞ lim 𝑠𝐺(𝑠) = ∞ (7.17)
𝑠→0
lim 𝐺(𝑠) = ∞ (7.13) To satisfy Eq. (7.17), 𝐺(𝑠) must take on the following form
𝑠→0
To satisfy Eq. (7.13), 𝐺(𝑠) must take on the following form 𝑠 + 𝑧1 ⋯
𝐺 𝑠 = 𝑛 → 𝑠 𝑛 𝑠 + 𝑝1 𝑠 + 𝑝2 ⋯ → 0 (7.14)
𝑠 + 𝑧1 ⋯ 𝑠 𝑠 + 𝑝1 ⋯
𝐺 𝑠 = 𝑛 → 𝑠 𝑛 𝑠 + 𝑝1 𝑠 + 𝑝2 ⋯ → 0 (7.14)
𝑠 𝑠 + 𝑝1 ⋯ 𝑛 ≥ 2: at least two integrations in the forward path
𝑛 ≥ 1: at least one pure integration in the forward path 𝑧1 𝑧2 ⋯
𝑛 = 1: lim 𝑠𝐺(𝑠) = ≠ ∞ → finite steady-state error
𝑧1 𝑧2 ⋯ 𝑠→0 𝑝1 𝑝2 ⋯
𝑛 = 0: lim 𝐺(𝑠) = ≠ ∞ → finite steady-state error → diverging ramps
𝑠→0 𝑝1 𝑝2 ⋯ 𝑛 = 0: lim 𝑠𝐺(𝑠) = 0
𝑠→0

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 17 Steady-State Error System Dynamics and Control 18 Steady-State Error

§2.Steady-State Error for Unity Feedback Systems §2.Steady-State Error for Unity Feedback Systems
Parabolic Input - Ex.7.2 Steady-State Errors for Systems with No Integrations
𝑠(1/𝑠 3) 1 Find the steady-state errors for inputs of 5𝑢(𝑡), 5𝑡𝑢(𝑡), 5𝑡 2 (𝑡) to
𝑒 ∞ = 𝑒parabola ∞ = lim = (7.20)
𝑠→0 1 + 𝐺(𝑠) lim 𝑠 2 𝐺(𝑠) the system 𝑅(𝑠) 𝐸(𝑠) 120(𝑠 + 2) 𝐶(𝑠)
𝑠→0 (𝑠 + 3)(𝑠 + 4)
To have zero steady-state error for a parabolic input
lim 𝑠 2 𝐺(𝑠) = ∞ (7.21) The function 𝑢(𝑡) is the unit step
𝑠→0 Solution
To satisfy Eq. (7.21), 𝐺(𝑠) must take on the following form
The closed-loop system is stable
𝑠 + 𝑧1 ⋯
𝐺 𝑠 = 𝑛 → 𝑠 𝑛 𝑠 + 𝑝1 𝑠 + 𝑝2 ⋯ → 0 (7.14) 1 5 5
𝑠 𝑠 + 𝑝1 ⋯ 5𝑢(𝑡) or 5/𝑠: 𝑒step ∞ = 5 × = =
1 + lim 𝐺(𝑠) 1 + 20 21
𝑛 ≥ 3: at least three integrations in the forward path 𝑠→0
𝑧1 𝑧2 ⋯ 1 5
𝑛 = 2: lim 𝑠 2 𝐺(𝑠) = ≠ ∞ → finite steady-state error 5𝑡𝑢(𝑡) or 5/𝑠 2: 𝑒ramp ∞ = 5 × = =∞
𝑠→0 𝑝1 𝑝2 ⋯ lim 𝑠𝐺(𝑠) 0
𝑠→0
2 → infinite steady-state error 1 10
𝑛 ≤ 1: lim 𝑠 𝐺 𝑠 = 0
𝑠→0 5𝑡 2 𝑢(𝑡) or 10/𝑠 3 : 𝑒parabola ∞ = 10 × = =∞
lim 𝑠 2 𝐺(𝑠) 0
𝑠→0
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

3
 18/03/2024

System Dynamics and Control 19 Steady-State Error System Dynamics and Control 20 Steady-State Error

§2.Steady-State Error for Unity Feedback Systems §2.Steady-State Error for Unity Feedback Systems
- Ex.7.3 Steady-State Errors for Systems with One Integration Skill-Assessment Ex.7.1
Find the steady-state errors for inputs of 5𝑢(𝑡), 5𝑡𝑢(𝑡), 5𝑡 2 (𝑡) to Problem A unity feedback system has the following forward TF
the system 𝑅(𝑠) 𝐸(𝑠) 100(𝑠 + 2)(𝑠 + 6) 𝐶(𝑠) 10(𝑠 + 20)(𝑠 + 30)
𝑠(𝑠 + 3)(𝑠 + 4)
𝐺 𝑠 =
𝑠(𝑠 + 25)(𝑠 + 35)
The function 𝑢(𝑡) is the unit step a.Find the steady-state error for the following inputs
Solution 15𝑢(𝑡), 15𝑡𝑢(𝑡), and 15𝑡 2 (𝑡)
The closed-loop system is stable b.Repeat for
1 5 10(𝑠 + 20)(𝑠 + 30)
5𝑢(𝑡) or 5/𝑠: 𝑒step ∞ = 5 × = =0 𝐺 𝑠 = 2
1 + lim 𝐺(𝑠) ∞ 𝑠 (𝑠 + 25)(𝑠 + 35)(𝑠 + 50)
𝑠→0
1 5 1
5𝑡𝑢(𝑡) or 5/𝑠 2: 𝑒ramp ∞ = 5 × = =
lim 𝑠𝐺(𝑠) 100 20
𝑠→0
1 10
5𝑡 2 𝑢(𝑡) or 10/𝑠 3 : 𝑒parabola ∞ = 10 × = =∞
lim 𝑠 2 𝐺(𝑠) 0
𝑠→0
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 21 Steady-State Error System Dynamics and Control 22 Steady-State Error

§2.Steady-State Error for Unity Feedback Systems §2.Steady-State Error for Unity Feedback Systems
Solution
a. First check stability b. First check stability
𝐺(𝑠) 10𝑠 2 + 500𝑠 + 6000 𝐺(𝑠)
𝑇 𝑠 = = 𝑇 𝑠 =
1 + 𝐺(𝑠) 𝑠 3 + 70𝑠 2 + 1375𝑠 + 6000 1 + 𝐺(𝑠)
10(𝑠 + 30)(𝑠 + 20) 10𝑠2 + 500𝑠 + 6000
= =
(𝑠 + 26.03)(𝑠 + 37.89)(𝑠 + 6.085) 𝑠5 + 110𝑠4 + 3875𝑠3 + 4.37 × 104𝑠2 + 500𝑠 + 6000
→ the closed-loop system is stable 10(𝑠 + 30)(𝑠 + 20)
=
5 5 (𝑠 + 50.01)(𝑠 + 35)(𝑠 + 25)(𝑠2 − 7.189×10−4𝑠 + 0.1372)
5𝑢(𝑡): 𝑒step ∞ = = =0
1 + lim 𝐺(𝑠) 1 + ∞ → From the 2nd-order term in the denominator, the system
𝑠→0
is unstable. Instability could also be determined using
5 5
5𝑡𝑢(𝑡): 𝑒ramp ∞ = = = 7.2917 the Routh-Hurwitz criteria on the denominator of 𝑇(𝑠)
lim 𝑠𝐺(𝑠) 10 × 20 × 30
𝑠→0 25 × 35 Since the system is unstable, calculations about
5 5 steady-state error cannot be made
5𝑡 2 𝑢(𝑡): 𝑒parabola ∞ = = =∞
lim 𝑠 2 𝐺(𝑠) 0
𝑠→0
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 23 Steady-State Error System Dynamics and Control 24 Steady-State Error

§3.Static Error Constants and System Type §3.Static Error Constants and System Type
Static Error Constants - Ex.7.4 Steady-State Error via Static Error Constants
- The relationships for steady-state error Evaluate the static error constants and find the expected error
1 for the standard step, ramp, and parabolic inputs
• For a step input, 𝑢(𝑡) 𝑒step ∞ =
1 + lim 𝐺(𝑠) 𝑅(𝑠) 𝐸(𝑠) 500(𝑠 + 2)(𝑠 + 5) 𝐶(𝑠)
𝑠→0
(𝑠 + 8)(𝑠 + 10)(𝑠 + 12)
1
• For a ramp input, 𝑡𝑢(𝑡) 𝑒ramp ∞ =
lim 𝑠𝐺(𝑠) 𝑅(𝑠) 𝐸(𝑠) 500(𝑠 + 2)(𝑠 + 5) 𝐶(𝑠)
𝑠→0
1 2 1 𝑠(𝑠 + 8)(𝑠 + 10)(𝑠 + 12)
• For a parabolic input, 𝑡 𝑢(𝑡) 𝑒parabola ∞ =
2 lim 𝑠 2 𝐺(𝑠) 𝑅(𝑠) 𝐸(𝑠) 500(𝑠 + 2)(𝑠 + 4)(𝑠 + 5)(𝑠 + 6)(𝑠 + 7) 𝐶(𝑠)
𝑠→0
- The limits static error constants 𝑠 2 (𝑠 + 8)(𝑠 + 10)(𝑠 + 12)

• Position constant, 𝐾𝑝 𝐾𝑝 = lim 𝐺(𝑠)

𝑠→0 Solution
• Velocity constant, 𝐾𝑣 𝐾𝑣 = lim 𝑠𝐺(𝑠)
𝑠→0 All closed-loop systems are indeed stable
• Acceleration constant, 𝐾𝑎 𝐾𝑎 = lim 𝑠 2 𝐺(𝑠)
𝑠→0
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

4
 18/03/2024

System Dynamics and Control 25 Steady-State Error System Dynamics and Control 26 Steady-State Error

§3.Static Error Constants and System Type §3.Static Error Constants and System Type
𝑅(𝑠) 𝐸(𝑠) 𝐶(𝑠) 𝑅(𝑠) 𝐸(𝑠) 𝐶(𝑠)
a. 500(𝑠 + 2)(𝑠 + 5) b. 500(𝑠 + 2)(𝑠 + 5)
(𝑠 + 8)(𝑠 + 10)(𝑠 + 12) 𝑠(𝑠 + 8)(𝑠 + 10)(𝑠 + 12)

- The limits static error constants - The limits static error constants
500 × 2 × 5 𝐾𝑝 = lim 𝐺(𝑠) = ∞
𝐾𝑝 = lim 𝐺(𝑠) = = 5.208 𝑠→0
𝑠→0 8 × 10 × 12 500 × 2 × 5 × 6
𝐾𝑣 = lim 𝑠𝐺(𝑠) = 0 𝐾𝑣 = lim 𝑠𝐺(𝑠) = = 31.25
𝑠→0 𝑠→0 8 × 10 × 12
𝐾𝑎 = lim 𝑠 2 𝐺(𝑠) = 0 2
𝐾𝑎 = lim 𝑠 𝐺(𝑠) = 0
𝑠→0 𝑠→0
- The steady-state error - The steady-state error
1 1 1 1
𝑒step ∞ = = = 0.161 𝑒step ∞ = = =0
1 + 𝐾𝑝 1 + 5.208 1 + 𝐾𝑝 1 + ∞
1 1 1 1
𝑒ramp ∞ = = =∞ 𝑒ramp ∞ = = = 0.032
𝐾𝑣 0 𝐾𝑣 31.25
1 1 1 1
𝑒parabola ∞ = = =∞ 𝑒parabola ∞ = = =∞
𝐾𝑎 0 𝐾𝑎 0
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 27 Steady-State Error System Dynamics and Control 28 Steady-State Error

§3.Static Error Constants and System Type §3.Static Error Constants and System Type
𝑅(𝑠) 𝐸(𝑠) 500(𝑠 + 2)(𝑠 + 4)(𝑠 + 5)(𝑠 + 6)(𝑠 + 7) 𝐶(𝑠)
c. Run ch7p1 in Appendix B
𝑠 2 (𝑠 + 8)(𝑠 + 10)(𝑠 + 12)
Learn how to use MATLAB to
- The limits static error constants • test the system for stability
𝐾𝑝 = lim 𝐺(𝑠) = ∞ • evaluate static error constants
𝑠→0
𝐾𝑣 = lim 𝑠𝐺(𝑠) = ∞ • calculate steady-state error
𝑠→0
500 × 2 × 4 × 5 × 6 × 7 • solve Ex.7.4 with System (b)
𝐾𝑎 = lim 𝑠 2 𝐺(𝑠) = = 875
𝑠→0 8 × 10 × 12
- The steady-state error
1 1
𝑒step ∞ = = =0
1 + 𝐾𝑝 1 + ∞
1 1
𝑒ramp ∞ = = =0
𝐾𝑣 ∞
1 1
𝑒parabola ∞ = = = 1.14 × 10−3
𝐾𝑎 875
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 29 Steady-State Error System Dynamics and Control 30 Steady-State Error

§3.Static Error Constants and System Type §3.Static Error Constants and System Type
System Type - Relationships between input, system type, static error
- The values of the static error constants, again, depend upon constants, and steady-state errors
Type 0 Type 1 Type 2
the form of 𝐺(𝑠), especially the number of pure integrations in Input
Steady-state
Static error Static error Static error
error formula Error Error Error
the forward path constant constant constant
1 1
- Given the system step, 𝑢(𝑡)
1 + 𝐾𝑝
𝐾𝑝 =constant
1 + 𝐾𝑝
𝐾𝑝 = ∞ 0 𝐾𝑝 = ∞ 0
𝑅(𝑠) 𝐸(𝑠) 𝐾(𝑠 + 𝑧 )(𝑠 + 𝑧 ) … 𝐶(𝑠) 1 1
1 2 ramp, 𝑡𝑢(𝑡) 𝐾𝑣 = 0 ∞ 𝐾𝑣 = constant 𝐾𝑣 = ∞ 0
𝑠 𝑛 (𝑠 + 𝑝1 )(𝑠 + 𝑝2 ) … 𝐾𝑣 𝐾𝑣
1 1 1
parabola, 𝑡 2 𝑢(𝑡) 𝐾𝑎 = 0 ∞ 𝐾𝑎 = 0 ∞ 𝐾𝑎 = constant
Feedback control system for defining system type 2 𝐾𝑎 𝐾𝑎

define system type to be the value of 𝑛 in the denominator or,

equivalently, the number of pure integrations in the forward path
• 𝑛 = 0 : type 0 system
• 𝑛 = 1 : type 1 system
• 𝑛 = 2 : type 2 system
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

5
 18/03/2024

System Dynamics and Control 31 Steady-State Error System Dynamics and Control 32 Steady-State Error

§3.Static Error Constants and System Type §3.Static Error Constants and System Type
Skill-Assessment Ex.7.2 a.
Problem A unity feedback system has the following forward TF The closed-loop transfer function
1000(𝑠 + 8) 𝐺(𝑠)
𝐺 𝑠 = 𝑇 𝑠 =
(𝑠 + 7)(𝑠 + 9) 1 + 𝐺(𝑠)
a.Evaluate system type, 𝐾𝑝 , 𝐾𝑣 , and 𝐾𝑎 1000(𝑠 + 8)
=
b.Use your answers to (a.) to find the steady-state 𝑠 + 9 𝑠 + 7 + 1000(𝑠 + 8)
errors for the standard step, ramp, and parabolic 1000(𝑠 + 8)
= 2
inputs 𝑠 + 1016𝑠 + 8063
Solution The system is stable The system is Type 0. Therefore
1000 × 8
𝐾𝑝 = lim 𝐺(𝑠) = = 127
𝑠→0 7×9
𝐾𝑣 = lim 𝑠𝐺(𝑠) = 0
𝑠→0
𝐾𝑎 = lim 𝑠 2 𝐺(𝑠) = 0
𝑠→0

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 33 Steady-State Error System Dynamics and Control 34 Steady-State Error

§3.Static Error Constants and System Type §3.Static Error Constants and System Type
b. TryIt 7.1
Use MATLAB, the Control
The steady-state error System Toolbox, and the
1 1 following statements to find 1000(𝑠 + 8)
𝑒step ∞ = = = 7.8 × 10−3 𝐾𝑝 ,𝑒step (∞), and the closed- 𝐺 𝑠 =
1 + 𝐾𝑝 1 + 127 (𝑠 + 7)(𝑠 + 9)
loop poles to check for
1 1 stability for the system of
𝑒ramp ∞ = = =∞
𝐾𝑣 0 Skill-Assessment Ex.7.2
1 1 numg=1000*[1 8];
𝑒parabola ∞ = = =∞
𝐾𝑎 0 deng=poly([-7 -9]);
G=tf(numg,deng);
Kp=dcgain(G)
estep=1/(1+Kp)
T=feedback(G,1);
poles=pole(T)

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 35 Steady-State Error System Dynamics and Control 36 Steady-State Error

§4.Steady-State Error Specifications §4.Steady-State Error Specifications

A robot used in the manufacturing of semiconductor random- - Specifications for a control system’s transient response
access memories (RAMs) similar to those in personal • damping ratio ζ
computers. Steady-state error is an important design • settling time 𝑇𝑠
consideration for assembly-line robots
• peak time 𝑇𝑝
• percent overshoot %𝑂𝑆
- Specification of a static error constant
• position constant 𝐾𝑝
• velocity constant 𝐾𝑣
• acceleration constant 𝐾𝑎

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

6
 18/03/2024

System Dynamics and Control 37 Steady-State Error System Dynamics and Control 38 Steady-State Error

§4.Steady-State Error Specifications §4.Steady-State Error Specifications

- For example, if a control system has the specification 𝐾𝑣 = - Ex.7.5 Interpreting the Steady-State Error Specification
1000, we can draw several conclusions What information is contained in the specification 𝐾𝑝 = 1000?
• The system is stable Solution
• The system is of Type 1 The system is stable
• A ramp input is the test signal The system is Type 0
• The steady-state error between the input ramp and the output The input test signal is a step
1 1 1
ramp is 1/𝐾𝑣 per unit of input slope The error per unit step is 𝑒step ∞ = = =
1 + 𝐾𝑝 1 + 1000 1001
Type 0 Type 1 Type 2 Type 0 Type 1 Type 2
Steady-state Steady-state
Input Static error Static error Static error Input Static error Static error Static error
error formula Error Error Error error formula Error Error Error
constant constant constant constant constant constant
1 1 1 1
step, 𝑢(𝑡) 𝐾𝑝 =constant 𝐾𝑝 = ∞ 0 𝐾𝑝 = ∞ 0 step, 𝑢(𝑡) 𝐾𝑝 =constant 𝐾𝑝 = ∞ 0 𝐾𝑝 = ∞ 0
1 + 𝐾𝑝 1 + 𝐾𝑝 1 + 𝐾𝑝 1 + 𝐾𝑝
1 1 1 1
ramp, 𝑡𝑢(𝑡) 𝐾𝑣 = 0 ∞ 𝐾𝑣 = constant 𝐾𝑣 = ∞ 0 ramp, 𝑡𝑢(𝑡) 𝐾𝑣 = 0 ∞ 𝐾𝑣 = constant 𝐾𝑣 = ∞ 0
𝐾𝑣 𝐾𝑣 𝐾𝑣 𝐾𝑣
1 1 1 1 1 1
parabola, 𝑡 2 𝑢(𝑡) 𝐾𝑎 = 0 ∞ 𝐾𝑎 = 0 ∞ 𝐾𝑎 = constant parabola, 𝑡 2 𝑢(𝑡) 𝐾𝑎 = 0 ∞ 𝐾𝑎 = 0 ∞ 𝐾𝑎 = constant
2 𝐾𝑎 𝐾𝑎 2 𝐾𝑎 𝐾𝑎

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 39 Steady-State Error System Dynamics and Control 40 Steady-State Error

§4.Steady-State Error Specifications §4.Steady-State Error Specifications

- Ex.7.6 Gain Design to Meet a Steady-State Error Specification
𝑅(𝑠) 𝐸(𝑠) 𝐾(𝑠 + 5) 𝐶(𝑠) Find the value of 𝐾 so that 𝑅(𝑠) 𝐸(𝑠) 𝐾(𝑠 + 5) 𝐶(𝑠)
𝑠(𝑠 + 6)(𝑠 + 7)(𝑠 + 8) there is 10% error in the 𝑠(𝑠 + 6)(𝑠 + 7)(𝑠 + 8)
steady state
Solution 1
Since the system is Type 1, the error stated in the problem 𝑒 ∞ = = 0.1
𝐾𝑣
must apply to a ramp input; only a ramp yields a finite error in a 𝐾×5
Type 1 system → 𝐾𝑣 = 10 = lim 𝑠𝐺 𝑠 =
𝑠→0 6×7×8
Input
Steady-state
Type 0
Static error
Type 1
Static error
Type 2
Static error
→ 𝐾 = 672
error formula Error Error Error
constant constant constant Applying the Routh-Hurwitz criterion, we see that the system is
1 1
step, 𝑢(𝑡)
1 + 𝐾𝑝
𝐾𝑝 =constant
1 + 𝐾𝑝
𝐾𝑝 = ∞ 0 𝐾𝑝 = ∞ 0 stable at this gain
ramp, 𝑡𝑢(𝑡)
1
𝐾𝑣 = 0 ∞ 𝐾𝑣 = constant
1
𝐾𝑣 = ∞ 0 Although this gain meets the criteria for steady-state error and
𝐾𝑣 𝐾𝑣
1 1
stability, it may not yield a desirable transient response
1
parabola, 𝑡 2 𝑢(𝑡) 𝐾𝑎 = 0 ∞ 𝐾𝑎 = 0 ∞ 𝐾𝑎 = constant
2 𝐾𝑎 𝐾𝑎

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 41 Steady-State Error System Dynamics and Control 42 Steady-State Error

§4.Steady-State Error Specifications §4.Steady-State Error Specifications

Run ch7p2 in Appendix B Skill-Assessment Ex.7.3
Learn how to use MATLAB to Problem A unity feedback system has the following forward TF
• find the gain to meet a steady-state error 𝐾(𝑠 + 12)
𝐺 𝑠 =
specification (𝑠 + 14)(𝑠 + 18)
• solves Ex.7.6 Find the value of 𝐾 to yield a 10% error in the steady
state
Solution The system is stable for positive 𝐾
For a step input
1
𝑒step ∞ = = 0.1
1 + 𝐾𝑝
12 × 𝐾
→ 𝐾𝑝 = 9 = lim 𝐺 𝑠 =
𝑠→0 14 × 18
→ 𝐾 = 189

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

7
 18/03/2024

System Dynamics and Control 43 Steady-State Error System Dynamics and Control 44 Steady-State Error

§4.Steady-State Error Specifications §5.Steady-State Error for Disturbances

TryIt 7.2 - Feedback control systems are used to compensate for
Use MATLAB, the Control 𝐾(𝑠 + 12) disturbances or unwanted inputs that enter a system, result that
System Toolbox, and the 𝐺 𝑠 =
following statements to solve (𝑠 + 14)(𝑠 + 18) regardless of these disturbances, the system can be designed
Skill-Assessment Ex.7.3 to follow the input with small or zero error
and check the resulting 𝐷(𝑠)
system for stability
𝑅(𝑠) 𝐸(𝑠) controller plant 𝐶(𝑠)
𝐺1 (𝑠) 𝐺2 (𝑠)
numg=[1 12];
deng=poly([-14 -18]);
G=tf(numg,deng); Feedback control system showing disturbance
Kpdk=dcgain(G); - Consider a feedback control system with a disturbance, 𝐷(𝑠),
estep=0.1;
K=(1/estep-1)/Kpdk
injected between the controller and the plant
T=feedback(G,1); 𝐶 𝑠 = 𝐸 𝑠 𝐺1 𝑠 𝐺2 𝑠 + 𝐷(𝑠) 𝐺2 (𝑠) (7.58)
poles=pole(T)
𝐶 𝑠 = 𝑅 𝑠 − 𝐸(𝑠) (7.59)
1 𝐺2 𝑠
→𝐸 𝑠 = 𝑅 𝑠 − 𝐷 𝑠 (7.60)
1 + 𝐺1 𝑠 𝐺2 𝑠 1 + 𝐺1 𝑠 𝐺2 𝑠
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 45 Steady-State Error System Dynamics and Control 46 Steady-State Error

§5.Steady-State Error for Disturbances §5.Steady-State Error for Disturbances

- To find the steady-state value of the error, apply the final value - Assume a step disturbance, 𝐷 𝑠 = 1/𝑠
theorem to Eq. (7.60) and obtain Substituting this value into the second term of Eq. (7.61),
𝑒 ∞ = lim 𝑠𝐸(𝑠) 𝑒𝐷 (∞) , the steady-state error component due to a step
𝑠→0
𝑠 𝑠𝐺2 𝑠 disturbance is found to be
= lim 𝑅 𝑠 − lim 𝐷 𝑠 𝑠𝐺2 𝑠 1 1
𝑠→0 1 + 𝐺1
𝑠 𝐺2 𝑠 𝑠→0 1 + 𝐺1 𝑠 𝐺2 𝑠 𝑒𝐷 ∞ = − lim =− (7.62)
= 𝑒𝑅 ∞ + 𝑒𝐷 (∞) 𝑠→0 1 + 𝐺1 𝑠 𝐺2 𝑠 𝑠 1
lim + lim 𝐺1 𝑠
𝑠→0 𝐺2 𝑠 𝑠→0
where,
• 𝑒𝑅 ∞ : the steady-state error due to 𝑅(𝑠) This equation shows that the steady-state error produced by a
𝑠 step disturbance can be reduced by increasing the dc gain of
𝑒𝑅 ∞ = lim 𝑅 𝑠 𝐺1 𝑠 or decreasing the dc gain of 𝐺2 𝑠
𝑠→0 1 + 𝐺1 𝑠 𝐺2 𝑠

• 𝑒𝐷 (∞) : the steady-state error due to the disturbance 𝐷(𝑠)

𝑠𝐺2 𝑠
𝑒𝐷 ∞ = − lim 𝐷 𝑠
𝑠→0 1 + 𝐺1 𝑠 𝐺2 𝑠
1 𝐺2 𝑠
𝐸 𝑠 = 1+𝐺 𝑠 𝐺2 𝑠
𝑅 𝑠 − 1+𝐺 𝑠 𝐺2 𝑠
𝐷 𝑠 (7.60)
1 1
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 47 Steady-State Error System Dynamics and Control 48 Steady-State Error

§5.Steady-State Error for Disturbances §5.Steady-State Error for Disturbances

- Rearrange the system so that the disturbance, 𝐷(𝑠), is depicted - Ex.7.7 Steady-State Error Due to Step Disturbance
as the input and the error, 𝐸(𝑠), as the output, with 𝑅 𝑠 = 0 Find the steady-state error component due to a step
𝐷(𝑠) disturbance for the system
𝑅(𝑠) 𝐸(𝑠) controller plant 𝐶(𝑠) 𝐷(𝑠) plant −𝐸(𝑠) 𝐷(𝑠)
𝐺1 (𝑠) 𝐺2 (𝑠) 𝐺2 (𝑠) controller 𝐺1 (𝑠) plant 𝐺2 (𝑠)
controller 𝑅(𝑠) 𝐸(𝑠) 1 𝐶(𝑠)
1000
𝐺1 (𝑠) 𝑠(𝑠 + 25)
Feedback control system Rearrange feedback system to show disturbance Solution
as input and error as output, with 𝑅(𝑠) = 0
The system is stable 1 1 1
To minimize the steady-state value of 𝐸(𝑠), we must either 𝑒𝐷 ∞ = − =− =−
1 0 + 1000 1000
• increase the dc gain of 𝐺1 (𝑠) so that a lower value of 𝐸(𝑠) will lim + lim 𝐺1 𝑠
𝑠→0 𝐺2 𝑠 𝑠→0
be fed back to match the steady-state value of 𝐷(𝑠), or The result shows that the steady-state error produced by the
plant
• decrease the dc value of 𝐺2 (𝑠), which then yields a smaller 𝐷(𝑠) −𝐸(𝑠) step disturbance is inversely proportional
𝐺2 (𝑠)
value of 𝑒(∞) as predicted by the feedback formula controller
to the dc gain of 𝐺1 (𝑠). The dc gain of
𝐺1 (𝑠) 𝐺2 (𝑠) is infinite in this example

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

8
 18/03/2024

System Dynamics and Control 49 Steady-State Error System Dynamics and Control 50 Steady-State Error

§5.Steady-State Error for Disturbances §6.Steady-State Error for Nonunity Feedback Systems
Skill-Assessment Ex.7.4 - A general feedback system, showing the input transducer,
Problem Evaluate the steady-state error component due to a 𝐺1 (𝑠), controller and plant, 𝐺2 (𝑠), and feedback, 𝐻1 (𝑠)
step disturbance for the system 𝑅(𝑠) 𝐸𝑎1 (𝑠) 𝐶(𝑠) 𝑅(𝑠) 𝐸𝑎 (𝑠) 𝐶(𝑠)
𝐺1 (𝑠) 𝐺2 (𝑠) 𝐺(𝑠)
𝐷(𝑠)
controller 𝐺1 (𝑠) plant 𝐺2 (𝑠)
𝑅(𝑠) 𝐸(𝑠) 𝑠+2 𝐶(𝑠) 𝐻1 (𝑠) 𝐻(𝑠)
1000
𝑠+4
- Pushing the input transducer to the right past the summing
junction yields the general nonunity feedback system, where
Solution The system is stable
𝐺 𝑠 = 𝐺1 (𝑠)𝐺2 (𝑠)
For a step input
𝐻 𝑠 = 𝐻1 (𝑠)/𝐺1 (𝑠)
1
𝑒𝐷 ∞ = − 𝐸𝑎 (𝑠) : actuating signal, 𝐸𝑎 𝑠 ≠ 𝐸 𝑠 = 𝐶 𝑠 − 𝑅(𝑠)
1
lim + lim 𝐺1 𝑠 If 𝑟(𝑡) and 𝑐(𝑡) have the same units, the steady-state error can
𝑠→0 𝐺2
𝑠 𝑠→0
1 be found, 𝑒 ∞ = 𝑟 ∞ − 𝑐(∞)
=
2 + 1000
= −9.98 × 10−4
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 51 Steady-State Error System Dynamics and Control 52 Steady-State Error

§6.Steady-State Error for Nonunity Feedback Systems §6.Steady-State Error for Nonunity Feedback Systems
𝑅(𝑠) 𝐸𝑎 (𝑠) 𝐶(𝑠) 𝑅(𝑠) 𝐸𝑎 (𝑠) 𝐶(𝑠) 𝑅(𝑠) 𝐸𝑎 (𝑠) 𝐶(𝑠) - Ex.7.8 Steady-State Error for Nonunity Feedback Systems
𝐺(𝑠) 𝐺(𝑠) 𝐺(𝑠)
Find the system type, the appropriate error constant associated
𝐻(𝑠) 𝐻(𝑠) 𝐻(𝑠) −1
with the system type, and the steady-state error for a unit step
−1
input. Assume input and output units are the same
𝑅(𝑠) 𝐸𝑎 (𝑠) 100 𝐶(𝑠) 𝑅(𝑠) 𝐸(𝑠) 𝐺(𝑠) 𝐶(𝑠)
𝑅(𝑠) 𝐸(𝑠) 𝐺(𝑠) 𝐶(𝑠)
𝑠(𝑠 + 10) 1 + 𝐺(𝑠)𝐻(𝑠) − 𝐺(𝑠)
1 + 𝐺(𝑠)𝐻(𝑠) − 𝐺(𝑠)
1
𝑠+5
- Form a unity feedback system by adding and subtracting unity Solution
feedback paths. This step requires that input and output units
Using the equivalent forward TF
be the same
100
- Combine 𝐻(𝑠) with the negative unity feedback 𝐺(𝑠) 𝑠(𝑠 + 10)
𝐺𝑒 𝑠 = =
- Combine the feedback system consisting of 𝐺(𝑠) and [𝐻 𝑠 − 1], 1 + 𝐺 𝑠 𝐻 𝑠 − 𝐺(𝑠) 1 + 100 1 100
−
leaving an equivalent forward path and a unity feedback 𝑠(𝑠 + 10) 𝑠 + 5 𝑠(𝑠 + 10)
Note: The final figure shows 𝐸 𝑠 = 𝑅 𝑠 − 𝐶(𝑠) explicitly 100(𝑠 + 5)
→ 𝐺𝑒 𝑠 =
𝑠 3 + 15𝑠 2 − 50𝑠 − 400
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 53 Steady-State Error System Dynamics and Control 54 Steady-State Error

§6.Steady-State Error for Nonunity Feedback Systems §6.Steady-State Error for Nonunity Feedback Systems
The equivalent forward TF TryIt 7.3 𝑅(𝑠) 𝐸𝑎 (𝑠) 100 𝐶(𝑠)

100(𝑠 + 5) Use MATLAB, the Control 𝑠(𝑠 + 10)

𝐺𝑒 𝑠 = 3 System Toolbox, and the
𝑠 + 15𝑠 2 − 50𝑠 − 400 following statements to find 1
𝑠+5
Thus, the system is Type 0, since there are no pure integrations 𝐺𝑒 (𝑠) in Ex.7.8

The appropriate static error constant is then 𝐾𝑝

G=zpk([],[0 -10],100);
100 × 5 5 H=zpk([],- 5,1);
𝐾𝑝 = lim 𝐺𝑒 𝑠 = =−
𝑠→0 −400 4 Ge=feedback(G,(H-1));
'Ge(s)'
The steady-state error
Ge=tf(Ge)
1 1 T=feedback(Ge,1);
𝑒(∞) = = = −4 'Poles of T(s)'
1 + 𝐾𝑝 1 − 5/4
pole(T)
The negative value for steady-state error implies that the output
step is larger than the input step

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

9
 18/03/2024

System Dynamics and Control 55 Steady-State Error System Dynamics and Control 56 Steady-State Error

§6.Steady-State Error for Nonunity Feedback Systems §6.Steady-State Error for Nonunity Feedback Systems
- Consider a nonunity feedback control system with disturbance lim(𝐺1 𝐺2 ) lim 𝐺2
𝑠→0 𝑠→0
𝐷(𝑠) 𝑒 ∞ = 1− − (7.70)
𝑅(𝑠) 𝐸(𝑠) controller plant 𝐶(𝑠) lim (1 + 𝐺1 𝐺2 𝐻) lim(1 + 𝐺1 𝐺2 𝐻)
𝐺1 (𝑠) 𝐺2 (𝑠) 𝑠→0 𝑠→0

𝐻(𝑠) For zero error

lim(𝐺1 𝐺2 ) lim 𝐺2
- The steady-state error for this system, 𝑒 ∞ = 𝑐(∞) − 𝑟(∞) 𝑠→0
= 1, 𝑠→0
=0 (7.71)
𝑒 ∞ = lim 𝑠𝐸(𝑠) lim(1 + 𝐺1 𝐺2 𝐻) lim(1 + 𝐺1 𝐺2 𝐻)
𝑠→0 𝑠→0 𝑠→0
𝐺1 𝐺2 𝐺2 The above two equations can always be satisfied if
= lim 𝑠 1− 𝑅− 𝐷 (7.69)
𝑠→0 1 + 𝐺1 𝐺2 𝐻 1 + 𝐺1 𝐺2 𝐻 • the system is stable
with step inputs and step disturbances, 𝑅 𝑠 = 𝐷 𝑠 = 1/𝑠 • 𝐺1 (𝑠) is a Type 1 system
lim(𝐺1 𝐺2 ) lim 𝐺2 • 𝐺2 (𝑠) is a Type 0 system
𝑠→0 𝑠→0
𝑒 ∞ = 1− − (7.70) • 𝐻(𝑠) is a Type 0 system with a dc gain of unity
lim (1 + 𝐺1 𝐺2 𝐻) lim(1 + 𝐺1 𝐺2 𝐻)
𝑠→0 𝑠→0

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 57 Steady-State Error System Dynamics and Control 58 Steady-State Error

§6.Steady-State Error for Nonunity Feedback Systems §6.Steady-State Error for Nonunity Feedback Systems
𝑅(𝑠) 𝐸𝑎1 (𝑠) 𝐶(𝑠) - Ex.7.9 Steady-State Actuating Signal for Nonunity Feedback Systems
𝐺1 (𝑠) 𝐺2 (𝑠)
𝑅(𝑠) 𝐸𝑎 (𝑠) 𝐶(𝑠)
100 Find the steady-state actuating signal
𝑠(𝑠 + 10)
𝐻1 (𝑠) for the system with a unit step input.
1
- The steady-state value of the actuating signal, 𝐸𝑎1 (𝑠), for a Repeat for a unit ramp input
𝑠+5
general feedback system Solution
𝑠𝑅(𝑠)𝐺1 (𝑠) 𝑠𝑅(𝑠)𝐺1 (𝑠) 𝑠𝑅(𝑠) × 1
𝑒𝑎1 ∞ = lim (7.72) 𝑒𝑎 ∞ = lim = lim
𝑠→0 1 + 𝐺2 (𝑠)𝐻1 (𝑠) 𝑠→0 1 + 𝐺2 (𝑠)𝐻1 (𝑠) 𝑠→0 100 1
1+ ×
𝑠(𝑠 + 10) 𝑠 + 5
Note: There is no restriction that the input and output units be 𝑠(1/𝑠) × 1
the same, since we are finding the steady-state For step input 𝑒𝑎 ∞ = lim =0
𝑠→0 100 1
difference between signals at the summing junction, 1+ ×
𝑠(𝑠 + 10) 𝑠 + 5
which do have the same units
𝑠(1/𝑠 2 ) × 1 1
For ramp input 𝑒𝑎 ∞ = lim =
𝑠→0 100 1 2
1+ ×
𝑠(𝑠 + 10) 𝑠 + 5

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 59 Steady-State Error System Dynamics and Control 60 Steady-State Error

§6.Steady-State Error for Nonunity Feedback Systems §6.Steady-State Error for Nonunity Feedback Systems
Skill-Assessment Ex.7.5 Solution The system is stable 𝑅(𝑠) 𝐸𝑎 (𝑠) 100 𝐶(𝑠)
𝑅(𝑠) 𝐸𝑎 (𝑠) 100 𝐶(𝑠) 𝑠+4
Problem Create a unity-feedback system
𝑠+4
1 1
1 𝐻𝑒 𝑠 = −1 𝑠+1
𝑠+1
𝑠+1
−𝑠 𝑅(𝑠) 𝐸𝑎 (𝑠) 100 𝐶(𝑠)
a.Find the steady-state error, 𝑒 ∞ = 𝑐(∞) − 𝑟(∞), for =
𝑠+1 𝑠+4
a unit step input given the non-unity feedback 𝐺(𝑠) −𝑠
system. Repeat for a unit ramp input. Assume input 𝐺𝑒 𝑠 = 𝑠+1
1 + 𝐺(𝑠)𝐻𝑒 (𝑠)
and output units are the same 100(𝑠 + 1) 𝑅(𝑠) 𝐸(𝑠) 100(𝑠 + 1) 𝐶(𝑠)
b.Find the steady-state actuating signal, 𝑒𝑎 (∞), for a = 2
𝑠 − 95𝑠 + 4 𝑠 2 − 95𝑠 + 4
unit step input given the nonunity feedback system.
Repeat for a unit ramp input

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

10
 18/03/2024

System Dynamics and Control 61 Steady-State Error System Dynamics and Control 62 Steady-State Error

§6.Steady-State Error for Nonunity Feedback Systems §6.Steady-State Error for Nonunity Feedback Systems
𝑅(𝑠) 𝐸(𝑠) 100(𝑠 + 1) 𝐶(𝑠) 𝑅(𝑠) 𝐸𝑎 (𝑠) 100 𝐶(𝑠)
𝑠 2 − 95𝑠 + 4 𝑠+4

1
a.The steady-state error, 𝑒 ∞ = 𝑐(∞) − 𝑟(∞) 𝑠+1
The system is Type 0, 𝐾𝑝 = lim 𝐺 𝑠 = 100/4 = 25 b.The steady-state actuating signal, 𝑒𝑎 (∞)
𝑠→0
𝑠𝑅(𝑠)𝐺1 (𝑠) 𝑠𝑅(𝑠) × 1
The steady-state error 𝑒𝑎 ∞ = lim = lim
𝑠→0 1 + 𝐺2 (𝑠)𝐻1(𝑠) 𝑠→0 100 1
𝑒step ∞ = 1/(1 + 𝐾𝑝 ) = 0.0385 1+ ×
𝑠+4 𝑠+1
𝑒ramp ∞ = ∞ 𝑠(1/𝑠) × 1 1
step input 𝑒𝑎 ∞ = lim = = 0.0385
Type 0 Type 1 Type 2 𝑠→0 100 1 104
Steady-state 1+ ×
Input
error formula
Static error
Error
Static error
Error
Static error
Error 𝑠+4 𝑠+1
constant constant constant
1 1 𝑠(1/𝑠2) × 1 1
step, 𝑢(𝑡) 𝐾𝑝 =constant 𝐾𝑝 = ∞ 0 𝐾𝑝 = ∞ 0 ramp input 𝑒𝑎 ∞ = lim = =∞
1 + 𝐾𝑝 1 + 𝐾𝑝 𝑠→0 100 1 0
1+ ×
ramp, 𝑡𝑢(𝑡)
1
𝐾𝑣 = 0 ∞ 𝐾𝑣 = constant
1
𝐾𝑣 = ∞ 0 𝑠+4 𝑠+1
𝐾𝑣 𝐾𝑣
1 1 1
parabola, 𝑡 2 𝑢(𝑡) 𝐾𝑎 = 0 ∞ 𝐾𝑎 = 0 ∞ 𝐾𝑎 = constant
2 𝐾𝑎 𝐾𝑎

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 63 Steady-State Error System Dynamics and Control 64 Steady-State Error

§7.Sensitivity §7.Sensitivity
Sensitivity: the ratio of the fractional change in the function to the - Ex.7.10 Sensitivity of a Closed-Loop Transfer Function
fractional change in the parameter as the fractional change of Calculate the sensitivity of the closed-loop transfer function to
the parameter approaches zero 𝑅(𝑠) 𝐸(𝑠) 𝐾 𝐶(𝑠) changes in the parameter 𝑎. How
Fractional change in the function,𝐹 𝑠(𝑠 + 𝑎) would you reduce the sensitivity?
𝑆𝐹:𝑃 = lim
∆𝑃→0 Fractional change in the parameter,𝑃 Solution
∆𝐹/𝐹
= lim The closed-loop transfer function
∆𝑃→0 ∆𝑃/𝑃 𝐾
𝑃∆𝐹 𝑇 𝑠 = 2
= lim 𝑠 + 𝑎𝑠 + 𝐾
∆𝑃→0 𝐹∆𝑃
The sensitivity
𝑃 𝛿𝐹
→ 𝑆𝐹:𝑃 = (7.75) 𝑎 𝛿𝑇 𝑎 −𝐾𝑠 −𝑎𝑠
𝐹 𝛿𝑃 𝑆𝑇:𝑎 = = =
𝑇 𝛿𝑎 𝐾 𝑠 2 + 𝑎𝑠 + 𝐾 2 𝑠 2 + 𝑎𝑠 + 𝐾
𝑠 2 + 𝑎𝑠 + 𝐾
An increase in 𝐾 reduces the sensitivity of the closed-loop
transfer function to changes in the parameter 𝑎
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 65 Steady-State Error System Dynamics and Control 66 Steady-State Error

§7.Sensitivity §7.Sensitivity
- Ex.7.11 Sensitivity of Steady-State Error with Ramp Input - Ex.7.12 Sensitivity of Steady-State Error with Step Input
Find the sensitivity of the steady-state error to changes in Find the sensitivity of the steady-state error to changes in
𝑅(𝑠) 𝐸(𝑠) 𝐾 𝐶(𝑠) parameter 𝐾 and parameter a 𝑅(𝑠) 𝐸(𝑠) 𝐾 𝐶(𝑠) parameter 𝐾 and parameter 𝑎
𝑠(𝑠 + 𝑎) with ramp inputs (𝑠 + 𝑎)(𝑠 + 𝑏) for the system with a step input
Solution Solution
The steady-state error for the system The steady-state error for this Type 0 system
𝑒 ∞ = 1/𝐾𝑣 = 𝑎/𝐾 1 1 𝑎𝑏
𝑒 ∞ = = =
The sensitivity of 𝑒(∞) to changes in parameter 𝑎 1 + 𝐾𝑝 1 + 𝐾 𝑎𝑏 + 𝐾
𝑎 𝛿𝑒 𝑎 1 𝑎𝑏
𝑆𝑒:𝑎 = = =1 The sensitivity of 𝑒(∞) to changes in parameter 𝑎
𝑒 𝛿𝑎 𝑎/𝐾 𝐾
𝑎 𝛿𝑒 𝑎 𝑎𝑏 + 𝐾 𝑏 − 𝑎𝑏 2 𝐾
The sensitivity of 𝑒(∞) to changes in parameter 𝐾 𝑆𝑒:𝑎 = = =
𝑒 𝛿𝑎 𝑎𝑏 (𝑎𝑏 + 𝐾)2 𝑎𝑏 + 𝐾
𝐾 𝛿𝑒 𝐾 −𝑎
𝑆𝑒:𝐾 = = = −1 𝑎𝑏 + 𝐾
𝑒 𝛿𝐾 𝑎/𝐾 𝐾 2
There is no reduction or increase in sensitivity
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

11
 18/03/2024

System Dynamics and Control 67 Steady-State Error System Dynamics and Control 68 Steady-State Error

§7.Sensitivity §7.Sensitivity
The steady-state error for this Type 0 system TryIt 7.4
𝑅(𝑠) 𝐸(𝑠) 𝐶(𝑠)
1 1 𝑎𝑏 Use MATLAB, the Symbolic 𝐾
𝑒 ∞ = = = Math Toolbox, and the (𝑠 + 𝑎)(𝑠 + 𝑏)
1 + 𝐾𝑝 1 + 𝐾 𝑎𝑏 + 𝐾 following statements to find
𝑎𝑏 𝑆𝑒:𝑎 in Ex.7.12
The sensitivity of 𝑒(∞) to changes in parameter 𝑎
𝑎 𝛿𝑒 𝑎 𝑎𝑏 + 𝐾 𝑏 − 𝑎𝑏 2 𝐾 syms K a b s
𝑆𝑒:𝑎 = = = G=K/((s+a)*(s+b));
𝑒 𝛿𝑎 𝑎𝑏 (𝑎𝑏 + 𝐾)2 𝑎𝑏 + 𝐾 Kp=subs(G,s,0);
𝑎𝑏 + 𝐾 e=1/(1+Kp);
The sensitivity of 𝑒(∞) to changes in parameter 𝐾 Sea=(a/e)*diff(e,a);
Sea=simple(Sea);
𝐾 𝛿𝑒 𝐾 −𝑎𝑏 −𝐾 'Sea'
𝑆𝑒:𝐾 = = =
𝑒 𝛿𝐾 𝑎𝑏 (𝑎𝑏 + 𝐾)2 𝑎𝑏 + 𝐾 pretty(Sea)
𝑎𝑏 + 𝐾
The sensitivity to changes in parameter 𝐾 and parameter 𝑎 is
less than unity for positive 𝑎 and 𝑏. Thus, feedback in this case
yields reduced sensitivity to variations in both parameters
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 69 Steady-State Error System Dynamics and Control 70 Steady-State Error

§7.Sensitivity §8.Steady-State Error for Systems in State Space

Skill-Assessment Ex.7.6 Analysis via Final Value Theorem
Problem Find the sensitivity of the steady-state error to changes - Consider the closed-loop system represented in state space
in 𝐾 𝑅(𝑠) 𝐸(𝑠) 𝐾(𝑠 + 7) 𝐶(𝑠) 𝒙ሶ = 𝑨𝒙 + 𝑩𝑟, 𝑦 = 𝑪𝒙 (7.84)
𝑠 2 + 2𝑠 + 10 - The Laplace transform of the error
Solution 𝐸 𝑠 = 𝑅 𝑠 − 𝑌(𝑠) (7.85)
The system is Type 0 𝑌 𝑠 = 𝑅 𝑠 𝑇(𝑠) (7.86)
𝐾𝑝 = lim 𝐺 𝑠 = 7𝐾/10 𝑇(𝑠) : the closed-loop transfer function
𝑠→0
The steady-state error for this Type 0 system → 𝐸 𝑠 = 𝑅 𝑠 [1 − 𝑇 𝑠 ] (7.87)
1 1 10 - Using Eq.(3.73) for 𝑇(𝑠)
𝑒 ∞ = = =
1 + 𝐾𝑝 1 + 7𝐾/10 10 + 7𝐾 𝐸 𝑠 = 𝑅 𝑠 1 − 𝑪 𝑠𝑰 − 𝑨 −1 𝑩 (7.88)
The sensitivity of 𝑒(∞) to changes in parameter 𝐾 - Applying the final value theorem
𝐾 𝛿𝑒 𝐾 −10 × 7 7𝐾 lim 𝑠𝐸(𝑠) = lim{𝑠𝑅 𝑠 1 − 𝑪 𝑠𝑰 − 𝑨 −1 𝑩 } (7.89)
𝑆𝑒:𝐾 = = =− 𝑠→0 𝑠→0
𝑒 𝛿𝐾 10 (10 + 7𝐾)2 10 + 7𝐾 𝑌(𝑠) −1 𝑩 +
𝑇 𝑠 = = 𝑪 𝑠𝑰 − 𝑨 𝑫 (3.73)
10 + 7𝐾 𝑅(𝑠)
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 71 Steady-State Error System Dynamics and Control 72 Steady-State Error

§8.Steady-State Error for Systems in State Space §8.Steady-State Error for Systems in State Space
- Ex.7.13 Steady-State Error Using the Final Value Theorem TryIt 7.5
Use MATLAB, the Symbolic −5 1 0 0
Evaluate the steady-state error for the system with unit step Math Toolbox, and the
and unit ramp inputs. Use the final value theorem following statements to find 𝑨 = 0 −2 1 , 𝑩 = 0 ,
−5 1 0 0 the steady-state error for a 20 −10 1 1
step input to the system of 𝑪 = −1 1 0
𝑨= 0 −2 1 , 𝑩 = 0 , 𝑪 = −1 1 0 Ex.7.13
Solution 20 −10 1 1
syms s
The steady-state error A= [-5 1 0; 0 -2 1; 20 -10 1];
𝑠+4 B=[0;0;1];
𝑒 ∞ = lim 𝑠𝑅(𝑠) 1 − C=[-1 1 0];
𝑠→0 𝑠 3 + 6𝑠 2 + 13𝑠 + 20 I=[1 0 0; 0 1 0; 0 0 1];
E=(1/s)*[1-C*[(s*I-A)^-1]*B];
𝑠 3 + 6𝑠 2 + 12𝑠 + 16 error=subs(s*E,s,0)
= lim 𝑠𝑅(𝑠) 3
𝑠→0 𝑠 + 6𝑠 2 + 13𝑠 + 20
For a unit step, 𝑅 𝑠 = 1/𝑠, and 𝑒 ∞ = 4/5. For a unit ramp,
𝑅 𝑠 = 1/𝑠 2, and 𝑒 ∞ = ∞. Notice that the system behaves
like a Type 0 system
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

12
 18/03/2024

System Dynamics and Control 73 Steady-State Error System Dynamics and Control 74 Steady-State Error

§8.Steady-State Error for Systems in State Space §8.Steady-State Error for Systems in State Space
Analysis via Input Substitution Ramp Inputs
Consider the closed-loop system represented in state space If the input is an unit ramp, 𝑟 = 𝑡, a steady-state solution, 𝒙𝑠𝑠
𝒙ሶ = 𝑨𝒙 + 𝑩𝑟, 𝑦 = 𝑪𝒙 (7.84) 𝒙𝑠𝑠 = 𝑉1 𝑡 + 𝑊1 𝑉2 𝑡 + 𝑊2 ⋯ 𝑉𝑛 𝑡 + 𝑊𝑛 𝑇
Step Inputs = 𝑽𝑡 + 𝑾, 𝑊𝑖 , 𝑉𝑖 are constants (7.97)
If the input is a unit step, 𝑟 = 1, a steady-state solution, 𝒙𝑠𝑠 𝒙ሶ 𝑠𝑠 = 𝑉1 𝑉2 ⋯ 𝑉𝑛 𝑇 = 𝑽 (7.98)
𝒙𝑠𝑠 = 𝑉1 𝑉2 ⋯ 𝑉𝑛 𝑇 = 𝑽, 𝑉𝑖 is constant (7.92) → 𝑽 = 𝑨 𝑽𝑡 + 𝑾 + 𝑩𝑡, 𝑦𝑠𝑠 = 𝑪 𝑽𝑡 + 𝑾 (7.99)
𝒙ሶ 𝑠𝑠 = 𝟎 (7.93) In order to balance Eq.(7.99)
Eq.(7.84) →𝟎 = 𝑨𝑽 + 𝑩, 𝑦𝑠𝑠 = 𝑪𝑽 (7.94) equate the matrix coefficients of 𝑡 𝑨𝑽 = −𝑩 or 𝑽 = −𝑨−1 𝑩
𝑽 = −𝑨−1 𝑩 (7.95) equating constant terms 𝑨𝑾 = 𝑽 or 𝑾 = 𝑨−1 𝑽
The steady-state error Substituting into (7.99) yields
𝑒 ∞ = 1 − 𝑦𝑠𝑠 = 1 − 𝑪𝑽 = 1 + 𝑪𝑨−1 𝑩 (7.96) 𝑦𝑠𝑠 = 𝑪 −𝑨−1𝑩𝑡 + 𝑨−1(−𝑨−1𝑩) = −𝑪 𝑨−1𝑩𝑡 + (𝑨−1)2𝑩)
The steady-state error
𝑒 ∞ = lim (𝑡 − 𝑦𝑠𝑠) = lim [ 1 + 𝑪𝑨−1𝑩 𝑡 + 𝑪(𝑨−1)2𝑩] (7.103)
𝑡→∞ 𝑡→∞
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 75 Steady-State Error System Dynamics and Control 76 Steady-State Error

§8.Steady-State Error for Systems in State Space §8.Steady-State Error for Systems in State Space
- Ex.7.14 Steady-State Error Using Input Substitution Skill-Assessment Ex.7.7
Evaluate the steady-state error for the system with unit step Problem Find the steady-state error for a step input using both
and unit ramp inputs. Use input substitution the final value theorem and input substitution methods
−5 1 0 0 𝑨=
0 1
,𝑩=
0
,𝑪= 1 1
𝑨= 0 −2 1 , 𝑩 = 0 , 𝑪 = −1 1 0 −3 −6 1
Solution 20 −10 1 1 Solution Using the final value theorem
−1
For a unit step input, the steady-state error 𝑒step ∞ = lim 𝑠𝑅 𝑠 1 − 𝑪 𝑠𝑰 − 𝑨 𝑩
𝑠→0
𝑒 ∞ = 1 + 𝑪𝑨−1 𝑩 𝑠 −1
−1
0
= 1 − 0.2 = lim 1 − 1 1
𝑠→0 3 𝑠+6 1
= 0.8
𝑠+6 1
For a ramp input, the steady-state error −3𝑠 𝑠 0
= lim 1 − 1 1
𝑒 ∞ = lim [ 1 + 𝑪𝑨−1 𝑩 𝑡 + 𝑪(𝑨−1 )2 𝑩] 𝑠→0 𝑠 2 + 6𝑠 + 3 1
𝑡→∞
= lim 0.8𝑡 + 0.08 𝑠 2 + 5𝑠 + 2
𝑡→∞ = lim = 2/3
𝑠→0 𝑠 2 + 6𝑠 + 3
=∞
HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

System Dynamics and Control 77 Steady-State Error

§8.Steady-State Error for Systems in State Space

Using input substituition

𝑒step ∞ = 1 + 𝑪𝑨−1 𝑩
−1
0 1 0
=1+ 1 1
−3 −6 1
−6 −1
3 0 0
=1+ 1 1
3 1
1
=1+ 1 1 −
3
0
2
=
3

HCM City Univ. of Technology, Faculty of Mechanical Engineering Nguyen Tan Tien

13