|
While many engineering undergraduate courses are structured
to introduce the theory related to various processes, students
with no industrial experience have little sense of how this
theory relates to real-life operations in the field. To address
this concern, we set out to create a learning activity dedicated
to help students relate theoretical concepts of process dynamics
and control presented in class to their practical application
in industry. Considering the prevailing trend towards a student-centred
approach to pedagogy, we wanted to provide students with a
basic simulation of system components. The program, designed
to enable students to visualise the relationship between input
variables and their effect on outputs, would engage the students
in bridging the theory and concepts to meaningful application
in practice. Hence, students could benefit from a learning
activity that focused on a visual representation of the critical
control elements in a simulated oil refinery furnace system.
This article examines the design decisions in the development
and evaluation of this learning activity, a web-based simulation
for process dynamics and control education, and discusses
the wider implications of online simulation interactivity
features.
Dubbed ‘SimFurnace’, the program was designed
to enable students to dynamically manipulate the inputs of
a crude oil furnace in the upstream end of a refinery. While
manipulating the fuel feed and feed flow rates of the system,
students are able to observe the system’s responses.
In the simulation freeform (practice) mode, students can also
see how the formula calculation changes dynamically with input
adjustments. Once students have had sufficient practice in
the freeform mode, they can access a set of assignments designed
to bring forth key process control concepts while problem-solving
common industrial challenges. The program, developed using
the Macromedia Flash MX authoring environment, is accessed
from the university web servers at the convenience of the
user. Srinivasan, Doiron, & Song (2003) offer a more in
depth description of SimFurnace.
Considerations for Design
Before creating a simulation, the rationale and the aims
of the project, the constraints and limitations in its development
and implementation, and a review of existing programs all
contribute to the initial graphic design and interactivity
specifications. For example, the screen features and user
input modes in SimFurnace are simple yet dynamic. Taking into
consideration the limited scope of refinery operations being
targeted and the fact that students were not expected to know
every detail of chemical plant systems, a high fidelity system
representation, such as those with complex and detailed displays
available for industrial training, was not deemed necessary
for meeting the learning objectives. However, emulating basic
system dynamics was very important.
After a review of existing online programs designed to help
students visualise process dynamics and control, we found
that many online simulations were still stand-alone applications
developed with Matlab/Simulink software. Applications created
with such software have three major weaknesses:
- Many programs process ‘batch’ type simulations
in which the parameters/variables are specified at the beginning
of the simulation run are not manipulated by the user during
the course of the simulation.
- The level of user interactivity with the model during
the learning exercise is minimal.
- The visualisation of the underlying process being simulated
is limited: most use only simple state equations and block-diagram
graphics to represent system elements and transfer functions.
In most cases, there is no dynamic visualisation of the
input/output relationship.
According to Fishwick (1995), these shortcomings are a serious
impediment to the effectiveness of simulations as a learning
tool. He emphasises that because computer simulations embody
the principle of “learning by doing”, even a geometric
model that looks good may not be satisfactory unless it has
graphical representation of the system dynamics. Furthermore,
these existing online simulations make little use of what
Ranky, Bengu & Spak (1997) call “anthropocentric
technologies” which, through intuitive interaction and
immediate feedback, enable the learner to explore and implement
concepts to a much deeper level.
Program Structure
A standard feature of simulation programs is a freeform
interaction with the simulated system. This enables students
to gain an intuitive sense of the interrelationships between
the variables in the simulated process. In SimFurnace, students
can direct the feed flow rate and the fuel flow rate into
the system while monitoring and controlling the product output
flow rate and temperature. The effects can be observed in
real-time as the product flow rate and temperature are charted
in the lower half of the screen. Students also have access
to information on the system’s underlying logic, a first-order-plus-dead-time
transfer function, and the ability to toggle the screen presentation
to reveal a dynamic flow diagram of the process. This freeform
section is also designed to familiarise the students with
the simulation environment that they will encounter in another
section: the assignment section (see Figure 1).
Although not all programs challenge the user to apply what
they practise in the freeform interaction with the simulation,
the design of assignment scenarios based on real-life problems
is of utmost importance. In SimFurnace, the objective of having
students complete the assignment section is to anchor their
learning by getting them to instinctively react to system
changes in a meaningful context. In doing so, students become
aware of the relationship between real-life issues such as
environmental factors and plant production requirements, and
how these aspects of production affect the system input and
output variables studied in class.
Figure 1: Assignment 4
Other Design Features
In order to provide feedback and some measure of control
for the instructor, the design of online simulations should
include a transparent tracking of user performance. As well
as controlling access to the program, the instructor needs
to enable and disable exercises, set new target ranges (parameters)
for the simulations and control exercises, and browse student
performance data and feedback comments. Hence an appropriate
Learning Management System (LMS) needs to be considered (see
Figure 2).
Figure 2: Custom Learning Management
Discussion
Taking into consideration what Yang & Alty (2002) describe
as the “need to develop the student’s intuition
for bridging the gap between theory and practice”, simple
web-based simulations like SimFurnace can be created with
existing off-the-shelf web authoring tools. Such customised
web-based simulations, with a specific context and focus,
can overcome the constraints of the classroom and enable students
to associate concepts in theory with an experience of their
meaningfulness in a setting related to their future profession.
Many new models for integrating online learning into the
curriculum are needed and their appropriateness in meeting
specific learning objectives in varied settings can only be
assured through a systemic approach to creating online learning
activities, designing meaningful interactivity, prototype
testing and learning outcome validation. Using SimFurnace
as an example, we have presented some of the design features
that we believe help students relate classroom theory to practice
in industry. Field trials investigating student learning gains
from using SimFurnace are underway and preliminary results
are encouraging.
References
Fishwick, P.A. (1995). Simulation Model Design and Execution:
Building Digital Worlds. Prentice Hall. http://www.cise.ufl.edu/~fishwick/book/book.html
(Last accessed 5 Jan 2004).
Ranky, P.G.; Bengu, G. & Spak, G.T. (1997). ‘The
Development and Application of Synchronous and Asynchronous
Technology Based Learning Aids for Undergraduate Engineering
Education’. Proceedings of the National Science
Foundation Engineering Education Innovators’ Conference,
Arlington, VA, USA.
Srinivasan, R.; Doiron, G. & Song, M. (June 2003). ‘Enhancing
Process Control Education using a Web-based Interactive Multimedia
Environment’. In Computer-aided Chemical Engineering
Vol. 15. Chen, B.Z. & Westerberg, A. (Eds.), Proceedings
of the 8th International Symposium on Process Systems Engineering.
Kunming, P.R.China. pp 1478–1483.
Yang, S.H. & Alty, J.L. (2002). ‘Designing a Multi-user
Web-based Distributed Simulator for Process Control Learning’.
In Grievnik, J. & van Schijndel, J. (Eds.), European
Symposium on Computer Aided Process Engineering–12.
The Hague, The Netherlands. pp. 1033–1038.
|
