Course detail

Intelligent controllers

FEKT-NIRGAcad. year: 2019/2020

Design and realisation of various continuous and discrete PID controllers. Optimization of adjusted parameters. Design, realisation and verification of simple adaptive controllers. Self-tuning controllers. Artificial intelligence in proces control (Fuzzy and neural controllers, Control of technological processes). Sensors, normalisation, connection, limited of disturbance. Individual project of simple self-tuning controller with artificial intelligence.

Language of instruction

English

Number of ECTS credits

6

Mode of study

Not applicable.

Learning outcomes of the course unit

Course absolvent should be an able to design and adjust process controller with standards algorithms. Also to design control algorithms with structures with principles artificial intelligence and implement them into a process computer too.

Prerequisites

The subject knowledge on the Bachelor´s degree level is requested.

Co-requisites

Not applicable.

Planned learning activities and teaching methods

Teaching methods depend on the type of course unit as specified in the article 7 of BUT Rules for Studies and Examinations.

Assesment methods and criteria linked to learning outcomes

Lesson. Max. 30 points.
Examination. Max. 70 points.

Course curriculum

Not applicable.

Work placements

Not applicable.

Aims

Target of this course is to acquaint students with practical design, realization and parameters setting of controllers with artificial intelligence for real process control. Detailed explanation of all aspects for usage of controllers in control loops. The student will design and verify the simple adaptive control algorithm as a semester project. Course absolvent should be able to design and to adjust process controllers with standard algorithms. He should be also prepared to apply control algorithms with new structures into a process computer.

Specification of controlled education, way of implementation and compensation for absences

The content and forms of instruction in the evaluated course are specified by a regulation issued by the lecturer responsible for the course and updated for every academic year.

Recommended optional programme components

Not applicable.

Prerequisites and corequisites

Not applicable.

Basic literature

PIVOŇKA, P. Číslicová řídicí technika, 151 s. Brno, 2003: 2003. s. 1 ( s.) (CS)

Recommended reading

Not applicable.

Classification of course in study plans

  • Programme EECC-MN Master's

    branch MN-SVE , 2 year of study, winter semester, elective specialised

Type of course unit

 

Lecture

39 hod., optionally

Teacher / Lecturer

Syllabus

Physical background of control.
Design and realisation of continuous PID controllers.
Different types of PID controllers, realisation, setting of parameters.
Design and realisation of discrete analogy of continuous PID algorithms.
Various types of controller's algorithm, realisation and comparison.
Philosophy of the process identification and design of controller's algorithm.
Optimum settings of controller's parameters, adaptive controllers, self tuning controllers, specific problems of adaptive control.
Artificial intelligence in controls algorithms.
Fuzzy logic.
Fuzzy controllers.
Artificial neural networks.
Neural controllers.
A/D and D/A converters, binary outputs and inputs, galvanic isolation, sensors and normalisation circuits, influence of disturbances.

Laboratory exercise

26 hod., compulsory

Teacher / Lecturer

Syllabus

Introductory lesson (organisation, instructions, safety). Demonstration.
Simulation in real-time in the program MATLAB.
Realisation of continuous PID controller, verification on the simulated model.
Discrete analogies of continuous PID algorithms, verification on the simulated model.
Verification of PID controllers on physical models. Anti-windup.
Various PID controllers, switching between algorithms.
Character of identification methods, last square method, semester project.
Programming and verification of last square method.
Design of PSD controller from identified parameters.
Fuzzy controllers. Fuzzy control of car.
Verification of self-tuning controller.
Neural controllers. Presentation of protocols, verification of programs.
Presentation of protocols, credit.