©2001 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Improving Laboratory Training for Automation and Process Control Courses with a Specifically Designed Testing Software Application

Felipe Mateos, Antonio M. López, Víctor M. González, Student Member, IEEE, and José M. Enguita

Abstract - A common problem in Automation and Process Control courses, often arises at the practical stage. Students have to implement and test their control programs using specialized control devices (such as microcontrollers or PLCs) and, in many cases, the use of scale models or real process components is not possible, having to deal with a large number of switches, LEDs, etc., which not only may be confusing, but also drastically reduces their motivation.

To mitigate this problem, other tools that could somewhat help by simulating the general behavior of the process (such as SCADA applications), are used. However, they are not intended for that purpose, so in most cases they are found inadequate.

PROSIMAX, an easy-to-use Windows based tool for testing control programs implemented on different control devices, is presented. Using PROSIMAX, a virtually infinite number of real processes are easily cloned into virtual processes which, once attached to the control device, will behave as the real ones. It is intended to minimize the main drawbacks of classical tools used for test in educational environments. However, PROSIMAX is intended to be used as an auxiliary tool, being known the fact that the use of real components or scale models is the best way to achieve a satisfactory educational result.

I. Introduction

In almost any course in Automation and Process Control, students have to deal with the design and test of control algorithms implemented in a variety of different control devices (microcontrollers, programmable logic controllers, personal computers,...). One of the most difficult and time-consuming tasks is the test of the control logic, which may involve up to 50% of the development time. Moreover, in most cases, it is quite difficult for the teacher to evaluate the correctness of the student's solutions.

Although the use of process models provide the students with an ideal environment for such tasks, the fact is that, often, they are not effectively used, mainly because of their high cost. In most cases, students are limited to test their control programs by means of debugging tools or large sets of switches and LEDs. This approach, although valid for small programs with a reduced number of input and output signals, completely fails for larger or more complex processes. There is a wide range of alternatives, but the tools available nowadays present some inconveniences, which make them difficult to use for teaching purposes. Powerful tools are expensive and difficult to use, while cheaper or easier-to-use tools usually present severe inconveniences, such as lack of flexibility.

This paper presents PROSIMAX^(TM) [1], an easy-to-use graphics-based software environment for testing the control logic, which tries to minimize the main drawbacks of the available tools, while still providing a good framework for both the students and the teacher. This tool is not intended to substitute for the use of real components, which is the best solution from the educational point of view, but in fact complements them, allowing the work on a larger number of different and more complex processes.

Section II of this paper provides a comparative study of classical tools used for these purposes based on their test capabilities. Section III provides an overview of PROSIMAX and Section IV presents an example of its use. Finally, in Sections V and VI, some results and a summary are presented.

II. Comparative Study

Classical tools usually employed in education to test control logic implementation can be classified into two main groups:

Hardware Tools:

Based on switches, potentiometers, signal generators and measurement devices, connected to the control device.
Simulation panels set up on charts representing the proccess to be controlled. Switches, potentiometers, LEDs, etc., which can easily be placed on the chart, allowing the user to follow the process evolution.
Ready-made scale models or prototypes, which provide a close approach to the real process.
Real process components, such as motors, contactors, valves, etc., arranged and connected to the control device as needed.

Software Tools:

Control equipment programming tools provided by the vendor, which usually include debugging capabilities.
Specific processes simulators which are software tools with a set of specific examples (configurable or not) that can be controlled by an external control device.
SCADA (Supervisory, Control and Data Adquisition) systems including highly graphic interface capabilites and script programming languages, which could be used for process simulation purposes.

To establish a comparision between these tools, a set of characteristics should be selected. This set may include:

Cost: average cost per workstation.
Learning Time: average time the user must spend in learning how to use the tool.
Closeness to the real process: how close the simulation is to the real process.
Development Time: average time invested before starting the test procedure, not including the learning time.
Flexibility: Ability of the tool to be used in a large number of different types of processes.
Connectivity: Ability of the tool to be connected to different control devices.
Maintenance: cost, in time and money, of keeping the tool working for long periods.
User Dependency: grade of user interaction needed to perform the test.

These characteristics can be represented on a graphic as the one shown in Fig. 1. Each characteristic is represented on one axis, and its optima value grows from the center of the graphic to the outermost extreme of the axis. By linking the evaluation points in each axis, a shape that can be used to extract a rule to evaluate the suitability of a tool can be created. The more regular and bigger the shape, the better the tool is.

Fig. 1. Comparison chart axes

Using these kinds of charts, a very intuitive and visual appreciation of the suitability of each tool can be achieved (Fig. 2), thus showing the main drawbacks of each one. It can even help the final user to choose the right tool for his/her necessities.

Fig. 2. Comparative charts

As can be seen in Fig. 2, all shapes of classical tools show striking irregularities representing their main drawbacks. However, PROSIMAX, satisfactorily covers all the evaluation characteristics, as can be seen in Fig. 3, resulting in a very suitable tool for testing medium size control systems in educational environments.

Fig. 3. PROSIMAX chart

III. PROSIMAX Overview

PROSIMAX [1, 2] is a software application for simulating industrial processes, the main objective of which is to provide the student with a flexible and intuitive tool for testing control programs and algorithms, implemented on programable logic controllers (PLCs), microcontrollers, microprocessors, and so on.

For the designer, PROSIMAX can reliably represent the behaviour of the process, and the programming bugs of the automatism can easily be detected, thus allowing exhaustive debugging, and reducing the time of the application development. It also provides an independence of the real system, useful for simulating the commissioning of an automation project as well as for the modification of the operation conditions without interrupting the production process.

For the teacher in Automation and Process Control, it means having a complete, simple and flexible set of tools allowing the configuration of a virtually infinite number of processes (plants) with very diverse elements for the student to work in.

PROSIMAX is built around two main components, both running under Microsoft Windows^(TM):

The edition environment: a set of CAD tools that allows the configuration of the behavior of the dynamic elements within the plant, as well as their connections. It is also possible to include static elements, e.g., identifying labels or colourful shapes, for aesthetic purposes.
The simulation environment: which guides the real-time evolution of plant elements and provides a graphical interface for the operator to follow the behavior of the controlled process. Dynamic objects generate the output events to the external control device based on the input commands received from it, the dynamic interaction with other objects and their internal parametrization defined at the edition stage.

Fig. 4. Testing the control program implemented on a Simatic S5 PLC with PROSIMAX

Edition Environment

Plants are constructed regardless of the control device that will be used. This independence is very useful in laboratories where there is a highly heterogeneous equipment, composed of controllers from many different vendors, thus allowing work on the same problem under different environments.

Object dynamic characteristics are configured at this stage. Objects can be easily placed in any screen position, redimensioned, configured and connected using a mouse. Fig. 5 shows a set of objects that make up a process in the edition environment.

Fig. 5. Edition environment.

Available objects, see Fig. 6, are grouped into four categories:

Liquid objects: vessels, pipes, valves, taps, etc.
Component objects: trucks, conveyor belts, etc.
Continuous material objects: trucks, conveyor belts, etc.
Miscellaneous objects: Sensors, displays, leds, buttons, keypads, etc.

Using this generic set of objects, a virtually infinite universe of industrial processes can be easily simulated.

Fig. 6. The object palette

Object dynamic behavior parametrization is performed by means of simple dialog boxes. Many objects allow a personalized design with the purpose of giving them an appearance as close to the real element as possible. For example, users may assign images created with external graphic applications, or digitalized with a scanner.

Connections between objects allows information and event interchange, without the necessity of using a programming language and in a very intuitive way.

Simulation Environment

In order to achieve the effective control of the simulated plant by the external equipment, it is necessary to map object input and/or output signals with their counterparts within the control device, as if physical wiring connections were being made.

The last step is just pressing the start button and concentrating on checking the operation of the implemented control algorithm.

IV. Working with PROSIMAX

This section shows how to build a plant from scratch, based on a real process specification, and how to simulate its evolution when controlled by a user-defined control program running in a Simatic S7-200 PLC.

Usually, the teacher would give his students some automation problem specification and a PROSIMAX plant representing it. The students can then concentrate their efforts on the development of the control algorithms and check their operation with the given plant.

Process Specification

A pump set like that shown in Fig. 7 takes water to a cistern from reserve tanks. The aim is to maintain the liquid in the vessel between a maximum and a minimum level.

The system consists of a switch where the mode of "Manual" or "Automatic" operation can be selected. When the system works on Manual mode (MAN position selected), the technician of the station controls the operation, Starting or Stopping the pump by means of the switch "Start/Stop" (SS). In Automatic mode (AUT position selected), the pump automatically starts or stops, depending on the level indicated by two digital detectors (MAX, level high and MIN, level low).

Fig. 7. The pumping set

1.- Edition Phase

A.- Object Identification

Each real object is associated with a PROSIMAX object, as shown in TABLE I.

TABLE I
Real - **PROSIMAX** Object Associations
Real Objects	PROSIMAX Objects	Comments
Manual/Automatic switch	1 Keyboard with two keys.	When one of them is active, the other one is disabled acting as a switch.
Start /Stop switch	1 Key	Usually open switch.
Cistern	Vessel	A vessel with a personalized shape in rectangular form. The capacity is defined as 1000 liters and initial content of 500.
Level Detectors	2 Digital content sensors	Created with a personalized design with the scheme of usually open contacts that close when the level of liquid reaches their sense threshold. Configured so that they sense at 20% and 80%, respectively.
Pump	Pump	Used for the introduction of liquid into the cistern. Configured with a flow rate of 70 units of volume per second.
Liquid Generator	Tap	Generates the necessary liquid for the circuit (it simulates the reserve tank). Configured with a flow rate of 70 units of volume per second.
Hydraulic Circuit	Pipes & Elbows	Conduct the vessel input/output flow. Used to give a more realistic effect.
Consumption	Valve	Consumption of water of the cistern. Not controlled by the PLC but by the user.
Level Display	Display Analog sensor	Vessel level display.
Pump Contactor	On/Off	Synchronised with the Pump object to allow the simulation of the connection and disconnection of the pump power circuit.

B.- Object Configuration

Each object in the plant must be parametrized in order to get the desired dynamic behavior. The configuration of an object is carried out by simple dialog boxes opened by clicking the right mouse button over its image. For example, Fig. 8 shows the dialog box for the configuration of the vessel. The vessel's capacity and the initial contents are set to correct values according to the process specifications

.
Fig. 8. Vessel configuration.

Some objects (e.g., the vessel) will not interact with the external control device, but others will (e.g., the pump). For those that do interact, it is very important to configure the input/output label (Fig. 9), with an easy name to remember, because it will be of great relevance in the communications configuration step, section 2.- B Object signal mapping.

Fig. 9. Input/output label configuration.

C.- Object Connection

Most objects need to be connected to other objects to specify the path the material will follow. For example, it is necessary to connect liquid objects to set up the different liquid circuits in a plant. In order to tell each object which circuit it belongs to, you must connect one object to another, starting from the first one, going on with the second, the third, and so on.

In this example, it is known that the liquid the tap generates should be sent to the cistern. It is necessary then to connect this tap to the cistern. The liquid of the cistern will also be emptied through the valve that simulates the consumption. So, the cistern has to be connected to the valve. A circuit formed by pipes and elbows could be included between the tap and the cistern to give the plant a more realistic appearance. These new objects will then be connected, in the direction that the liquid will flow, as has already been shown.

Some objects do not interact with any other in the plant, so they don`t need to be connected because they will only receive or send information to the external control device.

In the following video, the process of selecting, placing, configuring and connecting an object is shown:

Click here to see the video

2.- Simulation Phase

A.- External Control Device Communication Driver Selection

Once the external control device has been chosen, the adequate communication driver must be selected, as shown in Fig. 10.

There are several communication drivers available in PROSIMAX at the moment:

By means of direct connection between the serial port of the PC and the programming port of the following PLCs: Simatic S5 90U, 95U, 100U, 115U and Simatic S7 200 from SIEMENS AG.
By means of some standard I/O data acquisition cards.
By means of a serial connection to another personal computer, one of the computers playing the role of the control device, and the other one being in charge of the process simulation. This connection is at an experimental stage.

However, new communication drivers could be developed to communicate with other external control devices, as these drivers are independent modules which can easily be installed into the simulation environment.

Fig. 10. External control device communication driver selection.

B.- Object signal mapping

Object input/output signals, previously defined by a simple label, as shown in section 1.- B Object Configuration, have to be mapped to their counterpart "physical signals" in the external control device (a PLC from the Siemens S7-200 series in this example). This step is done within the specific driver configuration dialog box, as it is independent from the plant construction (see Fig. 11). Table II and Table III show the mapping used for the pump example objects.

Fig. 11. Object signal mapping.

TABLE II
**PROSIMAX** Output Signals (PLC Physical Input Signal)
PROSIMAX Output Signals	Description	PLC Physical Input Signal
MAN	Manual/Automatic switch: Manual mode	E 0.0
AUT	Manual/Automatic switch: Automatic mode	E 0.1
SS	Start /Stop switch	E 0.2
MIN	Level Detectors: Minimum level	E 0.3
MAX	Level Detectors: Maximum level	E 0.4

TABLE III
**PROSIMAX** Input Signals (PLC Physical Output Signals)
PROSIMAX Input Signals	Description	PLC Physical Output Signals
KMB	Pump	A 0.0
PKT	Pump Contactor	A 0.0

C.- Process simulation and robustness tests

By pressing the start button in the simulation environment, the process will start evolving in real time and the student can watch how the control program keeps the vessel's level between the minimum and maximum. Also, he/she can follow the process evolution, showing liquid flow through the pipes when it is the case, sensor activation depending on the vessel's level, pump start when commanded by the PLC, etc., as shown in the following video:

Click here to see the video

Robustness of the control logic can be tested by means of introducing disturbances in the simulated process while the external control device is guiding it. This can be done by clicking with the mouse on the objects, which will react, changing their internal state. For example, a tap that is closed will open after clicking and vice-versa. The more disturbances the control logic is able to handle, the more robust it is.

V. Results

PROSIMAX is being successfully used in the University of Oviedo [3, 4]. It has mitigated equipment deficiencies derived from the high number of different centers where Automation and Control courses are taught. Nowadays, different courses are taught in several different campuses spread over our region, involving more than 600 students distributed in 6 different careers. For this reason, to set up a unique fully equipped laboratory would be useless, and having as many fully equipped laboratories as needed is being difficult due to lack of resources.

Before the use of PROSIMAX, students had to test their work mainly using equipment based on switches and LEDs, which restricted the size and complexity of planned problems. Nowadays, the use of PROSIMAX has allowed more complex problems to be faced, improving students' formation of understanding. The next list shows some of the practical work developed by students in a recent course:

Bottle filling and labeling machine automation.
Temperature control in a manufacturing area.
Edifice automatic lighting.
Automatic length measurement system of plate in a steel manufacturing line.
H.I.C. quality control test automation.
Basket game machine control.
Fountain water effects automation.
Coffee machine automation.
Car cruising speed control system.
Greenhouse automation.
Mine elevator control.

The opinion of teachers and students about practical work in courses has also radically changed: teachers can suggest more complete problems to solve, and students, by following the results of their work in a graphical and more realistic way, face the problem as a challenge, instead of a tedious task.

PROSIMAX has also been requested from many academic centers in Spain, including universities, high schools, and technical formation centers. Even high school teachers are trained at institutional centers, such as the CFID (Center for the Training, Development, and Innovation in Industrial Mechanics), using PROSIMAX.

Important private companies, such as ACERALIA S. A., use the tool to train their staff up.

Finally, it has also been requested and used by professional engineering companies both for testing their control programs and for trade exhibitions, although it is not specifically oriented to professional development environments.

VI. Summary and Future Directions

Using classical methods for testing control programs running in control devices presents serious problems in educational environments. To mitigate these drawbacks, a software tool has been developed allowing the simulation of a virtually infinite number of processes, without the need of programming and in a very easy and intuitive way. PROSIMAX directly communicates with the control device, receiving input signals and generating output events. PROSIMAX helps in detecting and correcting critical errors within the control program that in any other way would have remained undetected. The process evolution can be easily followed by means of a graphical interface, increasing the motivation of the students.

Nevertheless, it must be stated that PROSIMAX is just an auxiliary tool, and that to achieve the best educational results, real equipment must be used so that the student acquires good skills in handling components involved in real processes.

PROSIMAX is currently being used with great success in the University of Oviedo, as well as in other universities, educational centers (public and private) and engineering companies. Based on this experience, it is possible to plan future enhancements and the inclusion of new tools to aid the students. This first approach was successful, however it would be possible not only to improve the tool shown in the present article, but to develop new tools to help in all the different stages involved in an automation project, mainly supervision and control program design.

Acknowledgments

The authors would like to thank the University of Oviedo, the Electrical, Electronics, Computers and Control Systems Engineering Department, and the Automation and Systems Engineering Area, for their help and support. Siemens has also supported the project from its beginning. Grants from FICYT (Foundation for the Foresting of Scientific and Applied Technology Research in Asturias) during three years made the development of this project possible.

References

[1] V.M. González, A.M. López, J.M. Enguita and F. Mateos. Prosimax 1.36. Industrial Process Simulator. User Manual., University of Oviedo, 1997

[2] V. M. González, F. Mateos, A. M. López, J. M. Enguita. "Diseño e Implementación de un Simulador de Procesos Industriales". In XIX Jornadas de Automática, pp. 245-251. Madrid, Spain.

[3] F. Mateos, V. M. González, M.A. Muñiz and J. L. Quintas. "Herramientas Software para Ayuda en el Diseño y Desarrollo de Proyectos de Automatización". In XIX Jornadas de Automática, pp. 415-419. Madrid, Spain, 1998.

[4] A. M. López, V. M. González, F. Mateos, J. M. Enguita, J. L. Pérez, M. A. Muñiz and J. A. Rodríguez. "Experiencias Docentes en Automatización Industrial Mediante el Uso de Herramientas Software para Simulación, Control y Supervisión de Procesos". In XX Jornadas de Automática, pp. 449-454. Salamanca, Spain, 1999.

[5] J. G. Bollinger and N. A. Duffie. Computer Control of Machines and Procesess. Addison-Wesley, 1989

[6] S. B. Morris. Automated Manufacturing Systems: Actuators, Contols, Sensors and Robotics. MacGraw-Hilll, 1994.

Author Contact Information

Felipe Mateos
Campus Universitario de Viesques, Ed. Departamental 2. 33204-Gijón, Asturias. Spain
Phone: +34 985 182 084
Fax: +34 985 182 068
E-mail: felipe@isa.uniovi.es

Antonio M. López
Campus Universitario de Viesques, Ed. Departamental 2. 33204-Gijón, Asturias. Spain
Phone: +34 985 182 663
Fax: +34 985 182 068
E-mail: antonio@isa.uniovi.es

Víctor M. González
Campus Universitario de Viesques, Ed. Departamental 2. 33204-Gijón, Asturias. Spain
Phone: +34 985 182 546
Fax: +34 985 182 068
E-mail: victor@isa.uniovi.es

José M. Enguita
Campus Universitario de Viesques, Ed. Departamental 2. 33204-Gijón, Asturias. Spain
Phone: +34 985 182 533
Fax: +34 985 182 068
E-mail: chema@isa.uniovi.es

Other links of interest:

University of Oviedo home page

Electrical, Electronics, Computers and Control Systems Engineering Department home page

Automation and Systems Engineering Area home page

GENIA (Integrated Automation Environment) home page

FICYT home page

Author Biographies

Felipe Mateos received the Ph.D. in Industrial Engineering from the University of Oviedo, Spain, in 1992. He is currently a lecturer, teaching courses in Automation and Control Engineering. He joined the Department of Electrical Engineering of the University of Oviedo in 1988 as an assitant lecturer. He has pursued since the beginning the development of hardware and software equipment with the aim of enhancing educational quality, being given several awards and obtaining patents for five of his products.

His research interests include distributed system integration for simulation, control and process supervision, and domotics.

Antonio M. López received the M.S. degree in Computer Science from the Escuela Técnica Superior de Ingenieros Industriales e Informáticos, Gijón, Spain, in 1996. He is currently studying for his Doctorate in the Department of Electrical Engineering of the University of Oviedo, Spain, where he has been working as an assistant lecturer since 1998, teaching courses in Automation and Control Engineering. He joined this department in 1993 with a grant for the development of PROSIMAX.

His research interests include simulation, soft computing technologies applied to automatic control, and dynamic system modeling.

Víctor M. González received the M.S. degree in Computer Science from the Escuela Técnica Superior de Ingenieros Industriales e Informáticos, Gijón, Spain, in 1997. He is currently studying for his Doctorate in the Department of Electrical Engineering of the University of Oviedo, Spain, where he has been working as an assistant lecturer since 1999, teaching courses in Digital Electronics and Control Engineering. He joined this department in 1993 with a grant for the development of PROSIMAX.

His research interests include simulation, process control and supervision and domotics.

José M. Enguita received the MS degree in Computer Science from the Escuela Técnica Superior de Ingenieros Industriales e Informáticos, Gijón, Spain, in 1997. He is currently studying for his Doctorate in the Department of Electrical Engineering of the University of Oviedo, Spain, where he has been working as an assistant lecturer since 1999, teaching courses in Automation and Control Engineering and Computer Networks. He joined this department in 1993 with a grant for the development of PROSIMAX.

His research interests include simulation, nondestructive inspection, data fusion, computer vision and conoscopic holography.

Return to Table of Contents