EE501T: ADVANCED CONTROL ENGINEERING (2020-2021)

Course Aims

To extend the work of earlier level courses in control systems to advanced, modern control methods, including transfer function, state vector and artificial-intelligence-based techniques, applicable to the design of both continuous and discrete-time systems.

Main Learning Outcomes

By the end of the course students should:

A) have knowledge and understanding of:

• classification of control systems
• the basic characteristics of discrete and hybrid control systems
• the effects of sampling on system behaviour
• design methods for discrete and hybrid systems
• various forms of state vector models for control systems and their use
• the analysis methods in state-space like: stability, controllability, observability, state feedback control, pole placement
• the characteristics and design of optimal controllers the concepts of state and parameter estimation
• experimental methods for system modeling
• basic self-tuning and adaptive control
• nonlinear control

B) have gained intellectual skills so that they are able to:

• select and apply methods for investigating the stability of continuous and discrete-time control systems
• select appropriate structures for feedback controllers, state observers and parameter estimators
• select and apply appropriate methods for digital controller design
• establish design procedures for basic optimal, self-tuning and adaptive controllers

C) have gained practical skills so that they are able to:

• set up, manipulate and use state-space, block diagram and transfer function representations of systems
• investigate stability, controllability and observability
• select an appropriate sample rate for the implementation of digital control
• apply the root locus and frequency domain methods to the design of digital controllers
• design state feedback controllers, state observers and parameter estimators
• devise appropriate experimental procedures and establish system models from time domain and frequency domain response data
• design basic optimal, self-tuning and adaptive controllers
• use Matlab-based software for control system analysis and design

D) have gained or improved transferable skills so that they are able to:

• use WWW-based material to aid learning
• engage in discussion regarding problem solving methodology
• devise checking procedures
• be flexible and multi-faceted in their thinking

Course Content

1. Introduction - system classification; continuous, discrete-time and hybrid systems; linear and non-linear systems; time invariant and time varying systems. Course philosophy.
2. State-space modelling ? solution of the state equation; Modeling in state space (electrical and mechanical systems), conversion between transfer function and state-space, eigenvalues and stability; eigenvectors and state matrix sensitivity, controllability and observability;
3. State-space control design - pole placement approach to state feedback design; output feedback, optimal control and Linear Quadratic Regulator; extension to non-linear systems.
4. System Identification - review of types and selection of system models; transfer function and state vector models; impulse and frequency response testing; time and frequency domain methods for identification from experimental data.
5. State and Parameter Estimation - structure and design of state observers and estimators; Luenberger observer design; RLS parameter estimators.
6. Discrete and Digital Control - mixed continuous and discrete time systems; z-transformation, z-domain transfer function and state-space model; system response and stability; stability analysis using Routh-Hurwitz; controller design using root locus; discrete approximations, frequency domain and direct methods.
7. Self-tuning and Adaptive Systems - system structures; gain scheduling controllers; self tuning controllers; model reference adaptive controllers.
8. Nonlinear systems definition; System stability analysis based on Lyapanov theory; Nonlinear controller and observer design based on nonlinear control techniques including sliding mode, feedback linearization, input-output linearization, and backstepping; Adaptive controller and observer introduction.

Information on contact teaching time is available from the course guide.

More Information about Week Numbers

2x homework assignment (set of problems) (30% each)

Design exercise (40%)

There are no assessments for this course.