1
University of CataniaComputer Engineering Department
Educational tools for complex topics: a case study for
Network Based Control Systems
Prof. Orazio MirabellaIng. Michele BrischettoIng. Antonino Raucea
2
Presentation Outline
Introduction Case Study - NBCS Learning Objectives The NBCS Simulator User Interface Learning with the simulator Feedbacks from the students
3
Introduction
Students in engineering courses, should develop the ability to correlate what they have learnt in different disciplines.
This is not always simple to accomplish because of:
● scarce cooperation between the teachers of the different courses ● limited time available to the students for each course.
This can prevent an unified vision of the various subjects studied and an insufficient ability to link them.
4
Case Study - NBCS
A Network Based Control System (NBCS) is a very challenging subject for engineers.
Different technological aspects should be taken into account in designing a NBCS. In our Faculty, these are dealt with in different courses:● Systems Theory ● Automatic Control● Process Control● Networks for Process Control
We have designed a multidisciplinary educational tool which aims to offer to the students a tool by which they can integrate skills acquired in those mentioned subjects to be applied in the study/design of a NBCS.
5
Digital ControlUsually students learn the theory regarding the analysis and synthesis of a closed loop system, and become able to define the continuous and discrete mathematical models of a controller, for example in the form of a PID controller, but in contrast, it is not given enough emphasis on the practical implementation.
Architecture of a Closed Loop Control SystemDigital control is a branch of control theory that uses digital computers to act as a system controller.
How a digital controller can be implemented?
6
NBCS
Delays and QoS(t) depend on different aspects of the network, such as its bandwidth,network congestion, network protocol used etc…All these aspects can decrease the quality of control and the controlled system could become unstable
How the network impacts on the Distributed Closed Loop System?
The use of a data network in a control loop has gained increasing attentions in recent years due to its cost effective and flexible applications
7
Learning Objectives
●To help students to learn what are the effects of a network over a NBCS.
●Can be used as a framework by which students can implement and test new network models to verify their effectiveness in this kind of applications.
●Can be used for the implementation and testing of specific techniques for distributed control.
8
Prerequisites and Educational Targets
Students should know:• the mathematical models of the most common systems;• the most typical input functions;• how to synthesize a continuous controller given some specification for the system to control;• how to pass from analog to discrete domain.
Educational Targets: how to code in a programming language the model of a dynamic system, for example with a finite difference model; how to implement the discrete controller algorithm in a programming language; how a network impacts over the remote controlled system; how to analyze the simulation results, and if it is required, how to modify the controller to take into account the network characteristics.
9
Transversal Cognitive Processes
1. System Modelling and Automatic Control: a. to know, or to be able to obtain, the mathematical model of a system b. to be able to synthesize a suitable controller to drive the dynamics of a system c. to be able to pass from analog to discrete models
2. Computer Networks: a. to know the main network architectures b. to be able to compare and valuate the characteristics of different network models for the different processes requirements.
3. Software Engineering and Programming: a. to be able to implement the discrete models of systems, controllers and networks, by mean of suitable languages and algorithms
10
Main User InterfaceMenu and Tool bars to configure the simulation to run
11
Main User InterfaceGraphical Animation related to the running system simulation
12
Main User InterfacePlotting area where input andsystem output curves are drawn
13
Main User InterfaceButtons to choose which curves in the diagram to hide or to show
14
Main User InterfaceInformations about the running simulation
15
Basic Learning with the toolFirst of all, students will find, already implemented, a suite of canonical transfer functions of different order for which suitable controllers have been computed and implemented in order to impose to the resulting closed loop system, some classical control theory specifications
Then it is possible to define the characteristics, in term of delay and loss probability, of the network which interconnects controller and remote system.
16
Other GUI Elements
17
Remote System BehaviourFirst students obtain a set of curves which reproduce the reference input signal and the system output
18
Remote System BehaviourThe output values computed at the previous step will be used to animate the graphical system model in such a way it follows the computed dynamic.
19
The architecture has been thought to allow a simple implementation and integration of new modules representing networks, systems, controllers and input reference functions.
The Software Architecture
20
The starting point is the analytical model of a system and a set of specifications regarding the desidered system dynamics.
•A system to study is choosen•The analitical model of the system is obtained•An analog controller able to impose the desired behaviour is synthetized•Passing from analog to discrete domain•The resulting discrete models have to be implemented as software modules using the proper classes•The network can be modeled and implemented as well•Simulation Phase•Critical analysis of the simulation results with different network configurations and controller parameters tuning
Advanced Learning
21
Students in the course of “Networks for Process Control” have been asked to use the tool.
Usually they did a set of simulations using the basic software modules
Some added new systems doing a more deep study
Both have been asked to write a short report about their work, simulation they did and the results obtained
Students activities
22
The tool has been appreciated especially for its simplicity, but also because it allowed to refresh and apply theoretical skills about different subjects in a simulated environment.
The tool has been said to be:• Easy to learn and use• Pleasant to experiment with it• Simple to upgrade adding new modules thanks to its modularity
Difficulties:• To obtain the system and controller analitical models
Students feedbacks
23
Thank you.
Questions?
