0% found this document useful (0 votes)
31 views13 pages

CS LabReport 9

The document describes an experiment conducted to: 1) Create a mathematical model of a QNET HVAC system and design a PI controller for temperature control. 2) Tasks included qualitatively testing the PI controller, examining effects of saturation and windup, and designing the PI controller to meet specifications of ξ=0.6 and ω0=0.125 rad/s, determining gains of Kp=24 V/°C and Ki=2.5 V/(°C s). 3) Post-lab tasks analyzed differences between on-off and PI control, and effects of varying proportional gain, integral gain, anti-windup parameter, and set-point weight on temperature response.

Uploaded by

Muhammad Afzaal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
31 views13 pages

CS LabReport 9

The document describes an experiment conducted to: 1) Create a mathematical model of a QNET HVAC system and design a PI controller for temperature control. 2) Tasks included qualitatively testing the PI controller, examining effects of saturation and windup, and designing the PI controller to meet specifications of ξ=0.6 and ω0=0.125 rad/s, determining gains of Kp=24 V/°C and Ki=2.5 V/(°C s). 3) Post-lab tasks analyzed differences between on-off and PI control, and effects of varying proportional gain, integral gain, anti-windup parameter, and set-point weight on temperature response.

Uploaded by

Muhammad Afzaal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

EEE-Control Systems

Lab #09

To Create the Mathematical Model of the QNET HVAC System and Design of
PI Controller for the System

Name Muhammad Afzaal

Registration Number FA18-BEE-093

Class FA18-BSEE-E

Instructor’s Name Sir Abubakar Talha Jalil


Lab #09: To Create the Mathematical Model of the QNET HVAC System and
Design of PI Controller for the System

Objectives
 To Create the Mathematical Model of the QNET HVAC System
 To Design a PI Controller on the System

Requirements
Hardware

 QNET Heating and Ventilation Trainer

Software

 Lab-View

Methodology
 The oscillations that occur with on-off control can be avoided by using a linear
proportional and integrating controller. To design such a controller analytically a simple
model representing the actual plant is needed. The block diagram of the closed loop
system is shown in Figure

 In lab-task1 for calibration Open the QNET_HVACT_PI_Control.vi. Make sure that the
correct device is chosen. Run the QNET_HVACT_On_Off_Control.vi. The cooling fan
is automatically activated when the Prototyping Board Power switch on the ELVIS unit
is ON. Let the actual temperature,, in the Temperature (C) scope settle until it stops
decreasing. Adjust the Temperature (C) scope scales to see both the reference and actual
temperatures. As illustrated in previous experiment, calibrate the temperature sensors by
clicking on the Calibrate button. This will align the chamber temperature, Tc, to the
measured ambient temperature, Ta. Activate the control by clicking on the Heater OFF.
 In lab-task 2 for Qualitative PI Control Ensure the QNET_HVACT_PI_Control.vi is
running. In the Signal Generator section set: 𝐴𝑚𝑝𝑙𝑖𝑡𝑢𝑑𝑒 = 0.5 ᵒ𝐶 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 = 0.02 𝐻𝑧
𝑂𝑓𝑓𝑠𝑒𝑡 = 1.5 ᵒ𝐶. In Control Parameters section set: 𝐾𝑝 = 4 𝑉/ ᵒ ,𝑖 = 0.5 𝑉/(ᵒ𝐶.𝑠),𝑏𝑠𝑝 = 1
,𝑇𝑟 = 1s . Examine the temperature response to the square wave input. Set 𝐾𝑖 𝑡𝑜 0
𝑉/(ᵒ𝐶.𝑠) and change the proportional gain between 2 𝑉/ ᵒ𝐶 and 10 𝑉/ ᵒ𝐶 Attach a
temperature response when using a low and high proportional gain.
 In lab-task 3 for saturation and windup In the Signal Generator section set: 𝐴𝑚𝑝𝑙𝑖𝑡𝑢𝑑𝑒 =
0.75 ᵒ 𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 = 0.02 𝐻𝑧,𝑂𝑓𝑓𝑠𝑒𝑡 = 1.5 ᵒ𝐶. In Control Parameters section set: 𝐾𝑝 = 8
𝑉/ ᵒ ,𝑖 = 4 𝑉/(ᵒ𝐶.𝑠),𝑏𝑠𝑝 = 1 ,𝑇𝑟 = 100s 4. Attach a response of the temperature and heater
voltage.
 In lab-task 4 In the Signal Generator section set: 𝐴𝑚𝑝𝑙𝑖𝑡𝑢𝑑𝑒 = 0.5 ᵒ𝐶 ,𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 =
0.02 𝐻𝑧 𝑂𝑓𝑓𝑠𝑒𝑡 = 1.5 ᵒ𝐶 . In Control Parameters section set:𝑝 = 8 𝑉/ ᵒ𝐶 , 𝐾𝑖 = 1 𝑉/(ᵒ𝐶.𝑠),
𝑏𝑠𝑝 = 0 ,𝑇𝑟 = 1s. Attach the temperature and heater voltage responses.
 In lab-task 5 design and test the PI controller.

Conclusion
In this lab we know about the QNET HVACT (Heating and Ventilation coach). The switch
regulator on QNET HVACT was using by lab View. Lab additionally assisted with finding out
about PI controller and how to design it in HVACT. With the help of hysteresis/delay system can
be controlled and worked under required conditions. The PI controller help with stabilize the
system. By reducing rise time, steady state error and overshoot the maximum and stabilize output
was taken from system which helped to improve the results.
In-Lab Tasks
Task-1: Qualitative PI Control
Set kp to 0.5 V/(◦C.s) and change the integral gain ki between 0.25 V/(◦C.s) and 2.0 V/(◦C.s) and
observe its effect on the temperature control performance.
Task-2: Saturation and Windup
Here is the graph of the chamber temperature and voltage.

Fig. 5 (Temperature of chamber)

Fig. 6 (Voltage of heater)


As the frequency of oscillating chamber temperature (red line) changes as we change the set-
point (blue line). So the heater stops and starts and different frequencies according to the
chamber temperature crossing the set-point. The cooling of chamber at higher temperature values
is faster and it rate of cooling gradually decreases as we move down to lower temperatures, that
is why due to faster cooling rates we get high frequency of oscillating chamber temperature.
What effect does increasing the anti-windup reset parameter have on the control signal and
on the temperature response?

Anti-windup forces systems to remain in saturation phase, means voltage has to decrease when
you start decreasing it but it won’t due to integral-windup. Setting the anti-windup reset time to a
value as large as 100 effectivelyshut off the anti-windup. Without anti-windupthe heater voltage
remains saturated at 8 V even after the temperaturehas reached its reference. This results in a
larger overshoot.

In the Control Parameters section, set Tr= 1.0 s.

Fig. 7 (Temperature of chamber)


Fig. 8 (Voltage of heater)
What effect does decreasing 𝑇𝑟have on the control signal and on the temperature response?

The temperature response and heater voltage used with anti-windup is depicted in fig.7 and fig.8.
As shown, anti-windup decreases the overshootin the temperature response. The heater voltage
begins to decrease as soon as the reference is reached.
Task-3: Set-Point Weight
Here is the graph ofthe chamber temperature and voltage.

Fig. 9 (Temperature of chamber)


Try the controller with a set-point weight of 1.

Fig. 11 (Temperature of chamber)

Fig. 12 (Voltage of heater)


Study what effects raising 𝑏𝑠𝑝has on the measured temperature signal in the Temperature
scope and the control signal shown in Voltage scope.
Increasing the set-point weight decreases the settling time of the response,i.e. reaches final value
faster, at the expense of more overshoot (and more control effort).
Task-4: PI Control According to Specifications
a) Designing the Controller
Find proportional and integral gains, i.e., 𝐾𝑝 𝑎𝑛𝑑 𝐾𝑖, needed for the response to satisfy the
following specifications:

 𝜉=0.60
 ⍵0=0.125 𝑟𝑎𝑑/𝑠

Results:

𝐾𝑣
𝑃(𝑠) =
𝑠

Δ𝑇𝑐
𝐾𝑣 =
Δ𝑡Δ𝑉ℎ

2 ξ ⍵0
𝑘𝑝 =
𝐾𝑣

⍵02
𝑘𝑖 =
𝐾𝑣

 kp= 24 V/oC
 ki= 2.5 V/(oC s)

b) Testing the Controller

Ensure the QNET_HVACT_PI_Control.vi is running and has been calibrated. In the Signal
Generator section set:

 𝐴𝑚𝑝𝑙𝑖𝑡𝑢𝑑𝑒 = 0.5 ᵒ𝐶
 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 = 0.02 𝐻𝑧
 𝑂𝑓𝑓𝑠𝑒𝑡 = 1.5 ᵒ𝐶
Enter the control gains calculated in part a in the Control Parameters Section of the VI.Examine
the measured temperature response using your designed PI gains. How is the performance of the
controller compared to the previous controller? Attach the temperature and heater voltage
responses

Result:

Fig. 14 (Temperature of chamber)


Fig. 15 (Voltage of heater)

The response with the PI is shown in above figure. The response behaves nicely. There is no
overshoot, little steady-stateerror, and the response reaches its set-point fairly quickly (i.e. low
settling time).
Post Lab Task
Task-1:

Give a comparative analysis of switch controller and PI controller for the temperature control
process.

Answer:

On-off control or relay feedback is one of the simplest control strategies. The switch is controlled
by certain conditions or limit points i.e temperature or voltage etc. Hystersis is added in the relay
switch to avoid rapid switching. It is a mathematical modeling of the process and strategy to
control the temperature or voltage.
On the other hand, PI controller is also used to control the switching of the system.The controller
aims to adjust the requirement of the system through input Vh. To make the process fast enough
as per our requirement we need to have very sharp value of the control input which might in turn
lead to damage the system. The proportional integral is a special case of the typical controller
family known as Proportional Integral Derivative (PID) controller.

Task-2:

Describe the effects of following parameters on the temperature response observed during the
experiment.
a) Proportional gain Kp
b) Integral gain Ki
c) Anti-Windup reset parameter Tr
d) Set-point Weight bsp

Answer:

a)If the VI was calibrated and ran properly, obtain aresponse similar to one of the plots. The
settling time is improved with a higher proportional gain and the steady-state error is decreased.

b) Increasing the integral gain makes the response faster, i.e., decreasesthe rise time. However,
increasing it can cause some overshoot due toactuator saturation.
c)Setting the anti-windup reset time to a value as large as 100 effectivelyshuts off the anti-
windup, without anti-windupthe heater voltage remains saturated at 8 V even after the
temperaturehas reached its reference. This results in a larger overshoot.

d)The response when the set-point weight is set to 1, bsp= 1Increasing the set-point weight
decreases the settling time of the response,i.e. reaches final value faster, at the expense of more
overshoot(and more control effort).

You might also like