THE PUBLISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Series A,

OF THE ROMANIAN ACADEMY Volume 6, Number 1/2005, pp. 000-000

COMPUTER AIDED PROCESS ENGINEERING

Paul Serban AGACHI

Babeş-Bolyai University, Faculty of Chemistry and Chemical Engineering

Computer Aided Process Engineering is one sector in the field of Process Engineering defining all
what is process related and computer assisted. It starts from the molecular modelling and passes
through computation of physical or chemical properties, mathematical modelling and simulation,
computer control, optimization and integrated process design. It is very applied engineering and
nowadays the name is redundant because there is no physically existent process or product in
elaboration which is not helped by computer if it is to be considered in the light of sustainable
development or to be valued on the market.
Key words: Computer Aided Process Engineering, Mathematical Modelling, Monitoring, Control,
Optimization, Process Design, Data Mining, Artificial Intelligence, Bio Informatics, E-Learning.

1. THE EVOLUTION OF PROCESS ENGINEERING

The Process Engineering (PE) is the field of industrial manufacturing or processing of whatever is
useful in the everyday life: production of potable water, pulp and paper, pharmaceutical drugs, polyethylene,
cars, steel, sulphuric acid, petrochemicals; production of energy, solvents, dyes, anticorrosive coating,
enzymes, pheromones, cement, textiles, food, beverages, waste processing and minimization etc. It covers
also the field of bio engineering.
Every scientific discipline has its characteristic set of problems and systematic methods for obtaining
their solutions - that is, its paradigm. The origins of PE are in 1888, when at Massachusetts Institute of
Technology, started a Chemical Engineering program, as an option to the program of Chemistry. First
chemical engineering paradigm is the unifying concept of unit operations proposed by Arthur D. Little in
1915. The concept became operational around 1920, just during the first petroleum crisis; other classical
tools of chemical engineering analysis were introduced: studies of the material and energy balance of
processes and fundamental thermodynamic studies of multi component systems. The second paradigm of the
field is the engineering science movement, consisting in phenomena in unit operations that are resolved at
the molecular level-micro scale, mechanistic models for these events, and mathematical models of processes.
The third chemical engineering paradigm expresses that nowadays, a confluence of intellectual advances,
technological challenges, and economic driving forces shape a new model of what chemical engineering is
and what chemical or already named process engineers will do: serve industries whose products are quickly
superseded in the marketplace by improved ones; serve industries that compete on the basis of quality and
product performance; expertise in the manufacturing of composite and structured materials from large
molecules; expertise in the manufacture of high-performance and specialty materials; expertise in designing
small-scale processes, batch processes; use more complete models, better approximations, and large
computers to solve problems rigorously; have multiple career changes. Academic research is also performed
by multidisciplinary groups of principal investigators; research and education also include the micro scale. In
all these fields, one can observe a new paradigm is occurring, that of the computational scientist.

__________________________________________________________
Reccomended by Mihai DRĂGĂNESCU, member of the Romanian Academy
 Paul Serban AGACHI 2

The forces behind this evolution are the explosion of new products and materials, the increased
competition for the world market and globalization, society’s increasing awareness of health risks and
environmental impacts.
In 1987, the National Research Council of the United States of America issued the so called
“Amundson” Report [1] which established the seven important directions in which the PE will develop in the
XXI century:
• Biotechnology and Biomedicine;
• Electronic, Photonic & Recording Materials and Devices;
• Polymers, Ceramics and Composites;
• Processing of Energy and Natural Resources;
• Environmental Protection, Process Safety and Hazardous Waste Management;
• Computer-Assisted Process and Control Engineering;
• Surfaces, Interfaces and Microstructures.
Thus, the foresighted “Amundson” Report envisioned the distinct field of what we consider
unavoidable today, Computer Aided Process Engineering (CAPE). More recent, one important workshop [2]
figuring out the advances in Information Technology described the changes in product engineering. The
synergies in chemical enterprise and information technology initiated unprecedented growth in the capability
and productivity.

2. COMPUTER AIDED PROCESS ENGINEERING

The development of CAPE found its way related closely to the development of the computing
hardware and the related software tools. Thus, the main directions of CAPE are nowadays:
• Modeling and Simulation of Unit Processes
• Studies on Flexibility, Operability and Dynamics of Processes
• Model-Based Manufacturing and Control
• Concurrent Process Engineering
• Whole Process Synthesis & Integration
All these developments take into consideration the recent affirmation of Sustainable Development.
Sustainable development was set on the agenda by the Brundtland Commission with the definition: ”meeting
the needs of the present without compromising the ability of future generations to meet their own needs”.
Figure 1 shows the consequences of this philosophy and the impact on CAPE.
It is obvious that a new approach supported by CAPE comes to be taken into consideration:
optimization, process improvement, process system design with sustainable options.
Modelling and Simulation in PE refers to molecular modelling [3], thermodynamic systems equilibria
[4], or macro systems modelling. One example is Computer Aided Drug Design, which is an important
contributor in the discovery and development of new pharmaceuticals. Special molecules are modelled and
then created to have properties in healing some specific diseases. To a large extent, the modelling and
simulation of the bio–processes taking place in the tissues or in the cell, are becoming more and more
popular in approaching. The hydrodynamics of the drugs circulation, the release of the active principles in
the body, the mass and heat transfer and chemical processes in the human body are studied largely [5],
creating the expectations that in some years (not more than 10) we shall be able to describe, calculate and
predict with high accuracy the effect of the drug on a certain specific target in the human body.
In Simulation and Computational Methods for Design and Operation, the predictions based on
modelling are good enough to obtain phase diagrams for real gases with accuracies exceeding experiments.
The Macro Modelling refers to the description of the steady state or dynamics of the industrial plants
which allows the implementation of improvements either in current operation, or in the structure of the plant
[6,7,8].
The Studies on Flexibility, Operability and Dynamics of Processes allow, through Sensitivity and
Controllability Analysis, the best operation mode of the plant, identification of the most important
parameters in the functioning of the plant. In the department of Chemical Engineering at the Faculty of
3 Computer Aided Process Engineering

Chemistry and Chemical Engineering in Cluj-Napoca there have been several researches in this field of
Sensitivity and Controllability Analysis [9,10,11].

Figure 1. Sustainable development and the consequences on CAPE

The classic control, using feed back approach, is still in the best position in the process industry,
regarding the number of applications, summing up around 80% [12,13]. There are also numerous new
applications of the classic control: development of specific control systems for new technologies: new
materials, bio technologies, de- pollution technologies; development of soft sensors [14] and better
exploitation of data mining technologies in obtaining synthetic information [15]; increase in the safety of
processes: detection, diagnosis and isolation of the faults –“fault-tolerant control”. At the same time,
advanced control is oriented towards the development of intelligent systems: expert systems, neural
networks, fuzzy techniques, genetic and evolutionary algorithms [16-18]. Industrial data monitoring and
processing are becoming more and more important in the management of the plants. An example is the
monitoring and data processing system implemented at UPSOM Ocna Mures [15]. The main panel of the
application, System of Energy Management is presented in the figure 2.
The system is working online since 1997, on a regular PC, in a very harsh industrial environment
without major problems. The system is gathering the data from the different transducers positioned on the
steam and other utilities production and consumption. It creates a historical archive of data on hourly,
monthly annual basis, calculates the real thermal load of the steam fluxes and gives important information on
the consumption in the different compartments of the soda ash factory. Based on this system, the users
reported an energy saving estimated, in average, at 15%. The system uses the LabView programming
environment
In more sophisticated situations, where the processes are either very sensitive or high energy
consumers, one of the latest control approaches is the Model Predictive Control. (Figure 3). At each step, the
manipulated variable is calculated in order to minimize the difference between the reference and the
predicted output of the controlled plant. The prediction on a certain horizon of time is based on an internal
model calculation.
Important researches in this field have been done in the Department of Chemical Engineering of
“Babeş-Bolyai” University of Cluj-Napoca [19 - 22]. A good example is the research on Fluid Catalytic
Cracking Unit (FCCU), which started in 1992 at California Institute of Technology, USA, based on the
request of Chevron Research and Development Department (Figure 4) and continued with the collaboration
 Paul Serban AGACHI 4

of SNC Petrom, Petrobrazi Refinery. The request of Chevron was to model and simulate another catalyst
circulation and an improved control scheme , with the scope of stabilizing the process.

Figure 2. Main panel of the Energy Management

COLOANĂ DE FRACŢIONARE

REACTOR

Past Future Reference

V14 6

REGENERATOR Abur V11

u(k+l) Eşapare
aer
1
Compresor gaze

Manipulated variable V7 3
svrgc Cuptor preîncălzire
materie primă

2
F3 Materie primă proaspătă
Suflantă aer de
V6 combustie

5
F4 F5 F2 F1
k k+1 k+2 k+l k+m k+p Aer
atmosferic Combustibil
Reciclu Motorină

Input horizon 1

2
Mărimi de proces:
Temperatură reactor
Output horizon 3
REGULATOR
Temperatură regenerator
4
PREDICTIV DUPĂ Zestre de catalizator reactor
5
MODEL Temperatură ciclon
6
Concentraţie oxigen în gaze arse

Figure 3. Model Predictive Control philosophy. Figure 4. Model Predictive Control of FCCU

The example of this approach is very interesting because at stake was an investment of 40 million USD
and the company wanted to know what results could be expected from the modifications. The results of the
work [23] showed that a significant improvement of the stability of the process could be reached only
through a modified procedure of operation and not through the important financial investment.
The research continued with the study of Predictive Control of another type of FCCU [24] working at
the refinery Brazi, in Romania. Nowadays the FCCU plant in the OMV company Brazi refinery, is computer
controlled and the basic control algorithms used are from the model based class. The control is hierarchical
at three levels: equipment, plant and refinery, the last level having an economic optimization objective.
Other orientations of CAPE are in the field of Process Optimization and Synthesis: mathematical
modeling and simulation of chemical processes using the integrated simulation and design software packages
ChemCAD, Aspen, Hysys, Pro/II, Matlab; validation studies of mathematical models, retrofit, energy
consumption reduction and optimization of chemical plants [25,26,27].
The last issue I would like to present in this paper on CAPE, is the e-learning or CAPE education.
Nowadays we are speaking a lot about sharing education through internet. E-learning covers many aspects
from e-library access to remote location laboratories [28,29]. Figure 5 sketches the organization of such a
laboratory. The laboratory is computer monitored and controlled. It can be supervised through video cameras
and microphones, the data, image and sound being transmitted to the remote location through Data-Voice
LAN and farther, through Internet, using a TCP/IP protocol. From the remote location it is possible to
measure parameters, to operate the pilot plant manually or automatically. In this way, the access to expensive
equipment is facilitated, creating a virtual Multiple User Laboratory.
All the software allowing the communication between the two locations was developed in the
Department of Chemical Engineering of “Babeş-Bolyai” University.
5 Computer Aided Process Engineering

Figure 5. Remote control laboratory

3. CONCLUSIONS

CAPE is nowadays recognized as a special branch of Process Engineering. The European Federation
of Chemical Engineering has a special CAPE Working Party which debates the developments of this chapter
of Process Engineering. As it could be observed from the paper, very much of the CAPE activity can be
identified as belonging to other fields of activity: Control Engineering, Biochemical Engineering, Medicine,
Information Technology, Environmental Engineering, Pharmaceutical Engineering, Chemistry and
Chemical Engineering. In fact, CAPE is a transdisciplinary field, overarching practically all the other
domains of application and its integrated approach create the premises of success of development of the field
in the next 50 years.

REFERENCES

1. *** - Frontiers in Chemical Engineering, Research, Needs and Opportunities, National Research Council, Washington, D.C., National
Academy Press, 1988
2. *** Information and Communications: Challenges for the Chemical Sciences in the 21st century, Workshop on Information and
Communications, Committee on Challenges for the Chemical Sciences in the 21st Century, National Research Council,
Washington D.C., National Academy Press, 2003.
3. RADA, S., SILAGHI-DUMITRESCU, I., A quantum chemical study on boron nitride and carbon nitride nanotubes,, CAPE Forum 2005,
25-26 February, Cluj -Napoca.
4. MARIA G., Relations between apparent and intrinsic kinetics of "programmable" drug release in human plasma, Chemical
Engineering Science, 60, p.1709-1723, 2005.
5. BATIU,I., Vapor-liquid equilibria in some binary systems containing terpenoids. Analysis in terms of group contributions model
(DISQUAC and Mod. UNIFAC Dortmund, CAPE Forum 2005, 25-26 February, Cluj-Napoca..
6. CRISTEA, V. M., AGACHI, S. P., Neural Networks Applications in Chemistry February, and Chemical Engineering, Cluj-Napoca
25-26 February, 2005,
7. ONICIU, L., AGACHI, S., BUGAN, J., IMRE, A., Steady State Model of the Electrochemical Process of Brine Electrolysis
Hungarian Journal of Chemical Engineering, 25, p. 81-89, 1997.
8. AGACHI, S., DICZIG, I., Modelarea matematica a procesului de fabricare a carburii de calciu , Rev. Chim.,41, no.4, p. 333-336,
1990.
9. AGACHI, S., CRISTEA, M., Controllability Analysis of a Model IV FCCU, 12-th International Congress of Chemical and Process
Engineering CHISA, Prague, p.10, 1996.
10. CRISTEA, M., AGACHI, S., GÖBEL,C., Sensitivity Analysis of an Industrial Dryer, 13th International Congress of Chemical
and Process Engineering CHISA, 23-28 August, Prague, 1998.
11. CRISTEA, V. M., AGACHI, Ş. P., Model Predictive Control of Inferred Variables and Dynamic Sensitivity Analysis Applied to
MPC Tuning, Buletinul Universitatii “Petrol-Gaze”, Ploiesti, Vol.L II, ST, no.1, p.52-57, 2000.
12. DUMITRESCU, S., AGACHI, P., S., Interaction of the control loops of a fuel cell 10 kW aggregate, Rev.Chim., 37, no.12,
p.1076-1079, 1986.
13. TOPAN, V., A., AGACHI, P., S., ONICIU, L., Chemical reaction control in a CSTR with nonlinear behaviour, based on the
mathematical model of the process, Rev.Chim., 31, no.1, p 53-56, 1980,
14. VELARDI, S.,A., BARRESI, A.,A., Modeling and monitoring of freeze drying process, CAPE Forum 2005, 25-26 February, Cluj -
Napoca.
 Paul Serban AGACHI 6

15. AGACHI, P., S., NAGY, Z., K., IMRE-LUCACI, A., CRISTEA, M.,V., SZASZ, G., Data acquisition and monitoring system of
an industrial power station, 13th International Congress of Chemical and Process Engineering CHISA, 23-28 August, Prague,
1998.
16. HAGAN, M.T., MENHAJ, M.H., Training feed forward networks with the Marquardt algorithm. IEEE Transaction on Neural
Networks, 5, no. 6, p.989-993,1994
17. CRISTEA, V. M., ROMAN, R., AGACHI, S. P., Neural Networks Based Model Predictive Control of the Drying Process,
European Symposium on Computer Aided Process Engineering (ESCAPE)-13, (Elsevier Science), Lappeenranta, Finland, 1-4
June, p.389-394, 2003.
18. BALDEA, M., CRISTEA, V. M., AGACHI, P. S., A Fuzzy Logic Approach to the Control of the Drying Process, Hungarian
Journal of Industrial Chemistry, 30, p.167-179, 2002.
19. NAGY, Z., AGACHI, S., Model Predictive Control of a PVC Batch Reactor, Computers & Chemical Engineering, no.6, p. 571-
591, 1997.
20. CRISTEA, V. M., MARINOIU, V., AGACHI, S. P., Reglarea predictivă după model a instalaţiei de cracare catalitică, Editura
Casa Cărţii de Ştiinţă, 183 p, Cluj-Napoca, 2003.
21. NAGY, Z., AGACHI, S., Model Based Control- Case studies, Editura Presa Universitara Clujeana, Cluj-Napoca, 2004.
22. CRISTEA, V. M., MARINOIU, V., AGACHI, S. P., Aspects of Traditional and Advanced Control of the Fluid Catalytic
Cracking Unit, Revue Roumaine de Chimie, 47, no.10-11, p.1127-1132, 2002.
23. MORARI, M., AGACHI, P., S., BOMBERGER, J., HUQ, I., Modeling and simulation of FCCU at Chevron Refinery El
Segundo, CA, Research Report for Chevron R&D Department, San Francisco, June 1992.
24. CRISTEA, M.,V.,. AGACHI, P., S., MARINOIU, M.,V., Simulation and Model Predictive Control of a UOP Fluid Catalytic
Cracking Unit, Chemical Engineering and Processing, Elsevier Science, 42, p.67-91, 2003.
25. CORMOS, C., AGACHI, S.P., Modeling and simulation of sodium beta-alaninate synthesis: comparison between commercial
softwares,16-th International Congress of Chemical and Process Engineering, Prague, Czech Republic, 22 – 26 August 2004
26. IMRE-LUCACI, A., SZASZ, G., NAGY, Z., K., KISS, A., AGACHI, P.,S., Dynamic Modeling of Continuous Distillation.
Comparison between Commercial Software, Proceedings of the 4th Conference on Process Integration, Modeling and
Optimization for Energy Saving and Pollution Reduction-PRES'01,Florence, May 20-23, p. 345-348, 2001.
27. IMRE-LUCACI, A., AGACHI, P., S., Optimizarea proceselor din industria chimică, Editura Tehnică Bucureşti, ISBN 973-31-
2113-4, 2002.
28. NAGY, Z. K., SIMON, L. L., CRISTEA, M. V., AGACHI, S. P., JIMENEZ, L., SCHAUFELBERGER, W., A Distillation Tele
laboratory for Distance CAPE Education, Proceedings of the 2003 WFEO/ASEE e-Conference, American Society for
Engineering Education, 2003.
29. NAGY, Z., K., AGACHI, P.,S., - Distance Learning applied in an Introductory Data Acquisition Course, the 5th International
Conference on Computer Aided Engineering Education, CAEE'99, September 22-24, 1999, Sofia, Bulgaria, pp. 250-254.

Received March 10, 2005