成人大片

Dorf and Bishop, Modern Control Systems, Chapters 5, 6, 7, 8, 9, 10, 12; Astrom and Hagglund, PID Controllers: Theory, Design and Tuning, Chapters 4 & 5

Maciejowski, Multivariable Feedback Design, Chapter 1; Xue, Chen and Atherton, Linear Feedback Control – Analysis and Design with Matlab, Chapter 6; Yu, Autotuning of PID Controllers, Chapters 2 and 3; Younkin, Industrial Servo Control Systems – Fundamentals and Applications.

As per university recommendations, it is expected that students spend 48hrs/week during teaching periods, and that a 3 unit course has a minimum workload of 156 hours regardless of the length of the course. Additional time may need to be spent acquiring assumed knowledge, working on assessment during non-teaching periods, and preparing for and attending examinations.

Below is a breakdown of the scheduled learning activities for this course:

Content:

Topic 1: Process Models (5%)

• Algebraic Systems
• Static models
• Dynamic models
• Integrator
• First Order Time Delay
• First order
• First Order Time Delay
• Second order
• Second Order Time Delay
• Identification
• Time domain
• Frequency domain

Topic 2: Frequency Domain (5%)

• Plotting frequency response function
• Bode Plots : Magnitude & Phase vs Frequency
• Nyquist - Complex Plane for G(jω)H(jω)
• Nichols Charts - Magnitude versus Phase Angle
• Frequency response of second order systems

Topic 3: Frequency Domain & Stability (5%)

• Review of Stability
• Routh-Hurwitz
• Root-Locus: s-plane
• The Cauchy Criterion
• The Nyquist Stability Criterion
• Gain, Phase & Delay Margins
• Effect of time delays on stability

Topic 4: Relationship between Open and Closed-Loop Frequency Response (5%)

• The basic relationship between open and closed-loop responses
• M and N Circles
• Maximum amplitude of closed-loop response
• Frequency at which this occurs (resonance frequency)
• Bandwidth

Topic 5: PID Control (5%)

• Various forms of Proportional-Integral-Derivative (PID) controllers
• The role of each elemental controller
• Actuator saturation and integral windup
• Proportional and derivative kick

Topic 6: PID Tuning - An Introduction (5%)

• Frequency domain modeling of simple process dynamics
• Ultimate gain
• Feature-based PID tuning techniques
• Ziegler-Nichols
• Chien, Hrones and Reswick
• Cohen-Coon

Topic 7: Advanced PID Tuning (5%)

• Pole Placement
• Dominant Poles
• Pole-Zero Cancellation

Topic 8: Sensitivity Relationships and Controller Design (5%)

• The sensitivity function
• The complementary sensitivity function
• The sensitivity of open loop and closed loop
• Loop shaping

Topic 9: More On Sensitivity Relationships and Robust Control (5%)

• Bode’s gain-phase relationship and integral theorem
• Relationship between sensitivity and proximity of frequency response function to critical point
• Relationship between sensitivity and phase and gain margins
• Robust stability criterion

Topic 10: New Tuning Methods - KT Tuning (5%)

• The issues with ZN tuning
• The need for alternative tuning methods
• Tuning to achieve a desired sensitivity
• Kappa-Tau tuning method

Topic 11: ITAE Optimal Systems (5%)

• Various time domain performance metrics
  • integral of the square of the error
  • integral of the absolute error
  • integral of the time multiplied by the absolute error
  • integral of the time multiplied by the square of the error
• Controller design using ITAE

Topic 12: Internal Model Control (5%)

• Internal Model Control (IMC)
• Advantages and disadvantages of model-based controllers
• Factorisation into invertible and non-invertible components
• The necessity for controllers to be proper

Topic 13: Servo Control (5%)

• Mechanical servo systems, including motors and amplifiers,
• transmission elements
• Backlash, its effect and techniques for reducing its influence
• Feedforward elements
• Dual loop control

Topic 14: Automatic Tuning (5%)

• The theory behind auto-tuning algorithms
• How to use auto-tuning algorithms
• The “relay method”

Matlab / Simulink modeling (30%)

• Design of a control system for a real physical system
• Hands-on implementation using Simulink and Space Control Desk

Text book: Nil

Assessment Rationale

Lectures, computer exercises, lab classes, and assignments are all intended to reinforce the learning outcomes of this course.

The computer exercises are designed to provide instruction of Matlab and Simulink while simultaneously developing the understanding of the students’ control knowledge through simulation.

The problem-based laboratory classes are intended to provide students with some practical experience in developing controllers for real (non-ideal) plants, as well as the use of instrumentation, and experience in report writing and communicating their results. Again, this assessment is intended to be formative and students receive written feedback on their work.

The examination is a summative assessment and is intended to assess the student’s knowledge and understanding of the course material and how it fits into the global engineering context.

All assignments and lab reports will be submitted electronically via MyUni.

Late assignments and reports will be penalised 10% per day. Extensions for assignments will only be given in exceptional circumstances and a case for this with supporting documentation can be made in writing after a lecture or via email to the lecturer. Hard copy assignments will be assessed and returned in 2 weeks of the due date. There will be no opportunities for re-submission of work of unacceptable standard. Due to the large size of the class feedback on assignments will be limited to in-class discussion resulting from questions from students.

All tutorials are submitted online using MyUni. Late tutorials cannot be accepted as they are submitted electronically via MyUni which automatically prevents late submission.