RMIT Classification: Trusted
MIET2006 Automatic Control
Lab Prac: Frequency Response and System
Identification
Student Name(s):
Student ID(s):
Lab Date:
Submission Due:
RMIT Classification: Trusted
1. Background
An input sinusoid Vm(t) can be characterized by its amplitude and frequency. When applying
an input sine wave to a cruise control system, the resulting output of the cruise control system
will be scaled and delayed sinusoid of the same frequency. For example, in Figure 1.1, t1
denotes the length of a period of the sinusoid and t2 shows the time delay between an input
voltage signal, Vm, and a scaled output speed signal, Ωm.
Figure 1.1: Period and phase delay of sinusoidal signals
Magnitude response analysis
The magnitude response of the resulting output changes with respect to the frequency of the
applied sinusoid and can be used to find the time constant τ of the cruise control system with
transfer function as shown in Equation 1.1. In particular, a Bode plot of the system’s
magnitude response similar to Figure 1.2 can be obtained by varying the frequency of the
applied sinusoid.
One of the characteristics of a Bode plot is the corner (cutoff) frequency ωc, which is defined
as the frequency where the gain of the system is 3 dB less than the maximum steady state
gain. An attenuation of 3 dB corresponds to the half power point of the system, where the
power at the cutoff frequency (in dB) is defined as 10 log 10 0.5 ≈ −3.01 𝑑𝑑𝑑𝑑. The
1 √2
corresponding magnitude at the cutoff frequency is then √2 ≈ 0.707, or 20 log 10 2 ≈
−3.01 𝑑𝑑𝑑𝑑, of the maximum system gain. The cutoff frequency is also referred to as the
bandwidth of a system and is a measure for how fast the system responds to a given input.
RMIT Classification: Trusted
Figure 1.2: Magnitude Bode plot
The first-order linear time-invariant (LTI) model of the cruise control has transfer function:
V𝑐𝑐𝑐𝑐 (𝑠𝑠) 𝐾𝐾 (1.1)
𝑇𝑇𝑇𝑇 = =
𝑉𝑉𝑟𝑟𝑟𝑟𝑟𝑟 (𝑠𝑠) 𝜏𝜏𝜏𝜏 + 1
where K is the steady state gain, τ is the time constant of the cruise control system, and
V𝑐𝑐𝑐𝑐 (𝑠𝑠) = 𝐿𝐿�𝑣𝑣𝑐𝑐𝑐𝑐 (𝑡𝑡)� and V𝑟𝑟𝑟𝑟𝑟𝑟 (𝑠𝑠) = 𝐿𝐿 �𝑣𝑣𝑟𝑟𝑟𝑟𝑟𝑟 (𝑡𝑡)� are the Laplace transforms of the speed after
cruise control and the reference speed, respectively. The time constant τ can be obtained by
applying sinusoidal inputs and observing the system response.
Substituting s = jω in Equation 1.1, the magnitude of the frequency response at a frequency ω
of the motor input voltage is then defined as
𝑉𝑉𝑐𝑐𝑐𝑐 (𝑗𝑗𝑗𝑗) 𝐾𝐾 (1.2)
|𝐺𝐺(𝑗𝑗𝑗𝑗)| = � �=� �
(
𝑉𝑉𝑟𝑟𝑟𝑟𝑟𝑟 𝑗𝑗𝑗𝑗 ) 𝜏𝜏𝜏𝜏𝜏𝜏 + 1
The magnitude or absolute value of a complex number z = x + jy is defined as
|𝑧𝑧| = |𝑥𝑥 + 𝑗𝑗𝑗𝑗| = �𝑥𝑥 2 + 𝑦𝑦 2 (1.3)
Therefore, determining the absolute value of the right side of Equation 1.2 subsequently leads
to the following expression for the system’s magnitude response with respect to the input
frequency:
𝐾𝐾 (1.4)
|𝐺𝐺(𝑗𝑗𝑗𝑗)| =
�(𝜏𝜏𝜏𝜏) 2 + 1
The system’s steady state (or low frequency) gain can then be obtained by setting ω = 0, i.e.
applying a constant signal:
𝐾𝐾 = |𝐺𝐺(0)| (1.5)
RMIT Classification: Trusted
2. In-lab exercise
In the first part of this experiment, you will determine the system gain K by applying a steady
state input. In the second part, you will use sinusoidal inputs to gain information about the
magnitude response of your system so that you can draw a Bode magnitude plot to determine
the time constant τ.
The cruise control system Simulink model shown in Figure 2.1 applies a sine wave and/or
constant reference value to the system and measures the corresponding speed on the cruise
control system.
Figure 2.1: Simulink model used to obtain the frequency response of the cruise control
system
Steady-state gain
As discussed in the background section, the steady state gain K of the system can be observed
by applying an input signal with a frequency of ω = 0 rad, see Equation 1.5.
1) Open the VirtualVehicle Simulink model.
2) To apply a constant value, set the Constant block to a value of 50 and the Gain block
to 0 (i.e. no sine wave).
3) Run the Simulink model. The response should look similar to Figure 2.2. Attach the
reference value and speed response you obtained.
4) Measure the speed of the system and calculate the steady state gain of the system, K.
RMIT Classification: Trusted
(a) Input: reference value
(b) Output: speed of the cruise control system
Figure 2.2: System response when applying a constant value 50 mph
RMIT Classification: Trusted
Magnitude response analysis
In this part of the lab, you will use sinusoidal inputs with different frequencies to determine
the time constant τ, of the system.
1) Given Equation 1.4, derive an expression to determine the time constant, τ .
Hint: Begin by evaluating the magnitude of the transfer function at the cut-off
frequency, ωc.
2) To configure the cruise control system model to apply a sine wave, set the Constant
block to 0 and the Gain block to 50. Set the Frequency in the Signal Generator block
to 0.01 rad/s.
3) Run the cruise control system. An example response is shown in Figure 2.3.
4) Measure the maximum positive speed of the response and compute the gain of the
system (linear and in dB). Enter the result in Table 2.1 below.
5) Repeat the previous steps for the remaining frequencies in Table 2.1. Enter the results
from the steady state analysis in the previous section in the line where the frequency
is 0.0 rad.
6) Using the plot command and the data you have collected in the previous steps and
summarized in Table 2.1, generate a Bode magnitude plot. The amplitude scale
should be in decibels (dB) and the frequency scale should be logarithmic.
Note: Ignore the entry for a frequency of 0 rad/s when drawing the Bode plot. The
logarithm of 0 is not defined.
RMIT Classification: Trusted
(a) Sine input (b) cruise control speed response
Figure 2.3: System response when applying a 50 mph 0.1rad/s sine wave
Example: Frequency = 0.1 rad/s
RMIT Classification: Trusted
Input Output Amplitude
frequency amplitude amplitude Amplitude ratio in dB
(rad/s) (mph) (m/s) ratio form
0.01 50 0.2
0.05 50 0.194
0.1 50 0.17
0.13 50 0.165
0.2 50 0.139
0.4 50 0.086
1 50 0.038
5 50 0.00772
20 50 0.00192
Table 2.1 Collect frequency response data
3. Calculation and analysis
Follow the above process, determine K and 𝜏𝜏, obtain transfer function of the control system.
4. Discussion
Theory background of this lab
Use Matlab plot Nyquist and Bode plot, discuss stability of the system.
Possible error(s)?
Application of system identification?
5. Conclusion
Conclusive result
6. feedback