SOME RECENT DEVELOPMENTS IN MODELLING AND

SIMULATION IN MANUFACTURING TECHNOLOGIES

MIKLÓS TISZA
University of Miskolc
Department of Mechanical Engineering, Miskolc-Hungary
tisza@gold.uni-miskolc.hu

Abstract. The application of modelling and simulation in manufacturing technologies is looking back
for more than twenty years old history. It covers a wide range of activities from the product design
through the process planning and process control including sophisticated methods of modelling and
simulation up to the implementation of fully computer integrated manufacturing. Among this
activities, the development of knowledge based systems has become an emerging field. Most of the
available knowledge based systems are based on simplified theoretical and empirical technological
rules. Therefore, they are usually not able to give accurate solutions and reliable predictions for
complicated manufacturing processes. In the last two decades significant progress have been also
achieved in the numerical modelling. In spite of the enormous development of hardware and software
facilities, the exclusive use of numerical modelling still seems to be veiy time- and cost consuming.
There fore, the integration of these two fields has gained great practical importance. In this paper, the
challenges and opportunities of the integration of knowledge based systems and finite element
modelling will be reviewed mainly from the viewpoint of synergy of research and industrial
applications.

Keywords: Manufacturing Technologies, Numerical Modelling, Simulation

1. Introduction

In the recent years the role of modelling and simulation in engineering and in
manufacturing industry have been continuously increasing. It is well known that during
manufacturing processes simultaneous effect of many different parameters can be observed.
This is the reason that in former years, detailed analysis of manufacturing processes could
have been done only with time-consuming and expensive trial and error methods. Due to
the recent developments both in the methods of modelling and simulation, as well as in the
computational facilities, modelling and simulation have become an everyday tool in
engineering practice. Besides the aforementioned facts, the emerging role of modelling and
simulation can also be explained by the growing globalisation and competition on the world
market requiring shorter lead times and more cost effective solutions.
Considering the available modelling techniques, two big groups may be distinguished:
commercially available, general purpose packages and sophisticated special task-oriented
packages. These latter ones may also be commercially available or developed by special
institutions or software houses for well-defined technological purposes.

2. Modelling and simulation techniques

There are many different packages available for simulation and modelling on the
market. Nowadays, finite element methods are most widely applied, but the finite volume
method and the boundary element technique are also often applied. In this paper, mainly the
finite element modelling and simulation will be analysed.

99
The fast development in computer techniques has made possible to simulate the
manufacturing processes in a complex way, by approximating more and more the physical
reality. There are two big groups of FEM codes: general purpose packages and
sophisticated process oriented packages. At the Department of Mechanical Engineering the
following packages are applied for studying manufacturing processes [1,2]:
• The SYSWELD™ package is mainly used to analyse phase transformation processes,
both in heat-treatment and in various welding processes;
• PAM-STAMP™ and AutoFORM™ are mainly applied for sheet metal forming
processes and the
• MARC™ and QFORM™ packages are used in bulk metal forming processes.

3. Modelling and simulation of technological processes

Technological processes are rather complicated: the optimum design can often be
achieved using time-consuming and very expensive experiments. Modelling and simulation
can provide very effective tools to decrease the development costs and to shorten the lead
time from the concept to realisation. In this paper, some selected examples of
manufacturing processes will be shown to illustrate how sophisticated computer techniques
can support design and production engineers in decision making.

First, a Computer Aided System elaborated for modelling and designing heat-treatment
processes will be described. The system elaborated at the Department of Mechanical
Engineering at the University of Miskolc is based on the non-equilibrium phase
transformation diagrams of steels [3]. Both isothermal and continuous cooling procedures
can be handled. The program package is suitable for die determination of technological
parameters for various heat-treatment processes. It can also be used to determine the
mechanical properties and the microstructural components achieved by the studied heat
treatment process. As an example, the determination of the technological parameters for
hardening of alloyed steels is shown in Figure 1.

Welding is another important technological process where the simulation has very
significant role. There are various computer aided packages available for the determination
of technological parameters of different welding procedures, to elaborate the so called
Welding Procedure Specifications (WPS), to study the phase transformation occurring
during the welding process both in the welded joint and in the heat affected zone, and to
perform various numerical simulations to predict the properties (including both the
mechanical properties and the microstructural changes) as a result of welding procedures.
With this technique various arc- and spot welding processes were studied. [4].

100
 Edzés Megeresztés
File Olaj ¥!z Megeresztés Vissza Nyomtatvány
•lominy szerinti anyagjel: 175 | Keies HRC
540
Issz MINOSEB TÍPUS ERTEK HRC11
•Ü17! 41 Ci 4 H MAX JL
171 41 Cr 4 H MIN
177 41 Cr 4 HH MAX T -> I
178 41 Ci 4 HH MIN 7 t -> T
3 T H
I ]a aE R j z : 259 Msz:
új E E a 2
HRC
R -> H
H -> R
i/R 1 8 .6 .4 .2 0 .2 .4 .6 .8 1
Az edzési keménység HRC edz: magban
56 52 46 46 42 42
T meg
37 32 29 29 26 26
L.mm 0 5 10 15 20 25 30 35 40 45 50 55 HRCedz A megetesztési kem. HRC meg: magban
HRC Hűtőközeg: olaj ÁI»éfBD HRCmeg 34 33 31 31 30 30
Elvát HRC a magban 28 26 25 25 24 24
^gy^fa l
Tmeg t meg C í Mn X Si X Ni * Ci X Mo X
540 7200 .415 75 1.05 0

Megeresztésí elválás
HRCedz 41.9 41 Ci4H
Rmeg | l 0 0 0 [HRCmeg 30 Kiqyuitég | TöHés

Figure 1. A Computer Aided System for the determination of the technological parameters
for heat-treatment processes

Sheet metal forming is one of the most widely applied manufacturing processes in the
manufacturing industry. Parts made from sheet metal can provide, with appropriate design,
a high strength to weight ratio. They are increasingly used from small electrical
components through the automobile industry up to the large aircraft structures for various
purposes. Despite the increasing number of applications of sheet metal parts, with the
exception of well-known handbooks, surprisingly little quantitative design information of a
general nature is available in the technical literature. Most companies use internal
guidelines for part design, based on experience with the geometries and materials used in
that specific company. While such design guidelines are extremely useful and practical,
they do not necessarily consider in detail the fundamental reasons for selecting a given
design. Thus, when a new part, a new material, or a new process is introduced the entire set
of experience-based design guidelines must be re-evaluated and modified. Therefore, it is
necessary to develop generic design methods based on metal forming analysis and on
systematic experimental investigation. This tendency can be clearly observed in the
development of various knowledge-based systems for designing sheet metal parts and for
process planning of forming procedures. As in many other metal forming applications,
process planning and design of dies for sheet forming can benefit from a combined
application of knowledge based systems and process modelling. Recently, many companies

101
are applying CAD/CAM techniques and knowledge-based expert systems to improve and
partially automate die design and manufacturing function.
Several program packages were elaborated for sheet metal forming processes at the
University of Miskolc at die Department of Mechanical Engineering. Among them first a
general CAD/CAPP/CAM system for the process planning of sheet forming processes
performed in progressive dies should be mentioned [6]. In this system, the process planning
and the die-design functions are integrated into a knowledge-based expert system. It
consists of a geometric module for creating, exporting and importing the object geometry, a
blank module for determining the optimum shape, size, and nesting of blanks, a
technological design module for designing the process sequence based on empirical rules
and technological parameters, a tool design module for designing the tools and selecting a
tool of standard size, and an NC/CNC post processor module for preparing programs for
NC/CNC manufacturing of tool elements.

There is an intense research activity all over the world to elaborate computer aided
engineering systems for supporting technological process planning in sheet metal forming.
Recently, there are two main approaches to achieve these goals. One of them is the
application of knowledge-based expert systems, which are generally based on simplified
plasticity theory and empirical technological rules. There are a great number of papers
dealing with the exclusive use of knowledge-based systems both in sheet and bulk metal
forming. But the exclusively knowledge based solutions have certain disadvantages: they
usually cannot provide an enough accurate solution to the problem since these systems are
generally based on simple technological rules with limited validity. Therefore knowledge-
based systems cannot predict for example the material flow, and usually cannot provide the
accurate stress and strain distribution inside the component [7].

As another approach, numerical techniques (recently mainly finite element modelling)

are applied for the analysis of the plastic deformation. The main objectives of the
application of numerical process simulation in metal forming are to determine appropriate
process parameters and to develop adequate die design by process simulation, to improve
part quality by predicting process limits and preventing flow induced defects. Besides
these, numerical process simulation also leads to reducing process and dies try-out, as well
as shorter lead times, while significantly reducing manufacturing costs. But the exclusive
use of numerical modelling has some drawbacks, too. In spite of the enormous development
of hardware and software facilities, the numerical modelling still seems to be very time-
and cost consuming and the reliability of results is often dependent on the experiences of
the user. It is partly due to the large number of operating parameters whose influence
should be investigated, and partly due to the numerical difficulties caused by the
complexity of the applied mathematical model to describe the material behaviour.

There fore, in the recent years the integration of these two fields (i.e. the knowledge-
based systems and numerical modelling) has gained primary importance. According to the
type of integration, two basic solutions can be found in die literature. The first group may
be regarded as a sequential integration. Since the knowledge-based systems are mainly
based on technological rules they are very appropriate to support the process engineer to
determine various alternative process plans and to analyse them from the technological

102
point of view. Using the knowledge-based systems as a first process-planning tool, the user
can easily determine technologically feasible process plans. The numerical methods then
can be used to supply a more deep and theoretically more accurate analysis of stresses and
strains, flow parameters, die-cavity filling and loads. In these solutions, the integration is
mainly limited to the process planning part of the product development cycle. Numerical
simulation and modelling of forming processes based on the knowledge of physics of
forming process and validated by experimental results and technological experiences, may
be regarded as the most powerful tool for optimising process plans and process parameters.
This integrated solution can significantly reduce - and in many cases eliminate - full-scale
process trials which was for a long time very characteristic for the metal forming industry.
By this virtual design and manufacturing both the development time and the costs can be
radically decreased comparing to conventional industrial practice.

The second group of integrated solutions rather fits into the field of simultaneous or
concurrent engineering. In this case, the numerical simulation is an inherent part of the
development cycle from the conceptual/product design through the process planning and
die design, and it is also present in the so-called virtual manufacturing phase. It is evident
that at the beginning stages of the development cycle (e.g. in the conceptual and product
design phase, as well as in the primary process planning stage), it is more practical to use
one-step simulation methods. It provides preliminary feedback such as a rough estimate of
the location of critical strains or wrinkling/fracture zones but not the absolute values. In the
later phases, if we wish to get more accurate results incremental simulation methods should
be used with correct modelling of materials, tools and forming parameters. On the basis of
the above considerations such kind of procedure appears to be the most suitable, since it is
able to overcome all the main drawbacks mentioned above. This integrated approach makes
possible to verify the results provided by an expert system, which is generally based on a
limited number of simple practical rules. If this approach is fully integrated into the
development cycle, the product designer and the process engineer can very accurately
model and determine the feasibility of the forming process in a very time- and cost-
effective way. As it clearly follows from the above-mentioned facts, the greatest benefit of
the integration of numerical modelling can be gained in the process planning stage. But the
numerical modelling can be applied through out the design phases as it was described
before.

A simulation and knowledge-based system (called DEEPEX) was developed for deep
drawing of axisymmetric components at the University of Miskolc [8], This system is based
on a feature-based, object oriented geometric description of the component to be drawn. In
this system, a two-stage process planning strategy is applied. In the first stage, the system is
capable to generate an initial - so-called geometrical - process sequence. This task is based
on the principles of the group technology, and is done by an automatic shape and feature
recognition processing. This is purely a geometry based first pass not considering any
technological constraints. It will serve as the input to the formability based second stage
that is used to refine this initial sequence. In the second stage of the process planning a
technologically feasible process plan is determined starting out of the geometrically
possible process plan and taking into consideration the technological - formability -
limitations. It means that the geometrically feasible process plan is tested and checked
against the technological constraints and rules stored in the knowledge and database of the

103
system. Obviously, this approach means that either more or less operations may be
technologically necessary than it was determined just from the geometry based feature
recognition. This is illustrated in Figure 2., where a geometrically feasible and
technologically acceptable process plan generated by the system is shown for the same deep
drawn component.
I iianwitju^i
USE
stisriíií
?
9RAM
»lt
IMjUiRV
L J -.una...
H06B,
ma
mm.:.
1 S ESOHi
SETTINGS
SJOTACHS
IKS:
ÜTÍUTV
•Wit-

IWluiuzM l.S M(f á'lli«lM l«M r •<--;'"!:: —

MitoC*4 Hm Mai.

Figure 2. Geometrical and technological process plans generated by

a knowledge based system
After both the geometrical and technological process plans have been completed, a
finite element analysis of each stage can be carried out to verify the feasibility of process
sequence. For this purpose the PAM-STAMP package is used. In this module the material
of the component to be drawn is handled as an anisotropic, elastoplastic shell material
(applying Hill's anisotropic plasticity algorithm) and various strain-hardening rules may be
taken into consideration as well. The tool materials in this analysis are taken as perfectly
rigid (so-called null material). The geometric description applied in this system can provide
the necessary input data for the discretisation of the component automatically. The applied
finite element program package also provides an easy way for adaptive remeshing as it is
required by the geometry of the component.

Recently, there are several efforts to develop systems for metal forming with the full
integration of knowledge-based solutions and FEM simulation. In these cases, the
simulation is integrated into the whole development process, i.e. the numerical simulation is
an inherent part of the development cycle from the conceptual product design, through the
process planning and die design, and it is also present in the so-called virtual manufacturing
phase. Therefore this kind of concept may be termed as Simulation and Knowledge Based
Design (SKBD). This concept provides a complete virtual design and manufacturing
environment [9].

104
 Since the production costs are dramatically increasing as the product development
cycle proceeds, it is of utmost importance to make the right decisions as early as possible,
líiis integrated approach of simulation, design and manufacturing can be regarded as the
most cost-effective solution in manufacturing engineering. Obviously, the final cost is
significantly influenced by the applied methods and particularly by the reliability of the
obtained results. Therefore, it seems to be very practical to use different simulation
techniques at different stages of product development cycle.

If we introduce simulation in the conceptual design stage, there are only very few real
data available for simulation. The purpose of the simulation at this stage to give a rough
estimation whether the required part can be manufactured or not. The so-called onestep
simulation techniques are ideal tools for this purpose. In these simulation codes the part
geometry, some of the important material and process parameters are required that are
usually available at this early stage. These codes are fast and accurate enough to indicate
the feasibility of the part for manufacturing. If the feasibility of the part is not proved it is
still the right time to modify the concept of the design, and if it is proved a more detailed
analysis can start.

Obviously, more accurate simulations can be performed by incremental methods and it

is also obvious that it needs more detailed information on the product, process parameters
and even the tool description is required to perform it. If the simulation is integrated with
knowledge based system that is capable to provide not only a practically proven process
plan, but also the necessary process parameters and tool details, all the required parameters
are available for a more reliable incremental simulation. The performance of an integrated
knowledge based simulation - having a good CAD interface - is of crucial importance to
get well optimised forming processes with the appropriate die constructions [10].

The simulation is still necessary in the prototyping and pre-production phases, too. It
will lead us to the so-called virtual manufacturing phase. If this simulation step is
neglected, the real feasibility of process plans and tool geometry can be verified when the
tools are already manufactured. Any modifications required in this case are rather costly
and tedious. If a systematic simulation is carried out in this stage, the necessary
modification still can be done with much less cost and work and the so-called die-tryout
time can be radically decreased [11].

4. Conclusions

In this paper, the challenges and the opportunities of application modelling and
simulation in materials science and manufacturing processes was described. On the basis of
the presented examples, it may be stated that the exclusive use of Knowledge Based
Systems and that of the Finite Element Simulation has certain drawbacks. If they are
integrated in the whole product development cycle - from the conceptual design, through
the process planning and die design, as well as in the manufacturing phase a complete
virtual design and manufacturing environment can be created in a joint Simulation and
Knowledge Based System.

105
 This concept provides significant advantages both in the design and in the
manufacturing phase. Applying die principles of Knowledge Based Expert Systems a fast
engineering solution can be achieved, which combining with the recent achievements in
FEM techniques provides theoretically more reliable results, i.e. more optimum design
throughout the development cycle. This integrated approach will lead to significantly
shorter lead times, better product quality and as a consequence more cost-effective design
and production.

5. Acknowledgements

This research was jointly supported by the Hungarian Academy of Sciences and the
Ministry of Education, which is gratefully acknowledged.

REFERENCES

1. TISZA, M. et.al.: Modellezés és szimuláció a mechanikai technológiákban,

Gépgyártástechnológia, 2000. No. 8. p. 13-18.
2. TISZA, M. et. al.: Numerikus modellezés az anyagtudományban és az
anyagtechnológiákban, Anyagok Világa, 2001. No. 10. p. 1-12.
3. GÁL, I., SCHAFFER, J.: Development of a Computer Aided System for Technological
Design of Heat-treatment Processes, MicroCAD"1998 Int Conf. 25-26. March 1998.
4. KIRK, C., TISZA, M., BALOGH, A., MILEHAM, A.: Comparison of Various
Numerical Methods Applied in Materials Processing Technologies, DAAAM" 99
Conference, Miskolc, 1999,
5. ADORJÁN, G.: Theoretical and Experimental Investigation of Welding Processes
Performed with Rotational Arc, PhD Thesis, University of Miskolc, 2000. p. 1-84.
6. TISZA, M.; RÁCZ, P.: A Complete CAD/CAM Package for Sheet Metal Forming, 2nd
Conf. on CAD/CAM and Robotics, 22-26. August 1990. London, p. 110-117.
7. TISZA, M.: Expert System for Sheet Metal Forming, J. of Materials Processing
Technology (1995) p. 423-432.
8. TISZA, M.: Application of FEM Simulation in Process Planning of Deep-drawing, 1st
ESAFORM Conference, Sophia Antipolis 17-20. March 1998. p. 169-178.
9. Hindman, C.; Ousterhut, K.: A Virtual Design System for Sheet Metal Forming J.
Materials Processing Technology v.84. (1998) p. 107-116.
10. TISZA, M.: Integration of Numerical Modelling and Knowledge Based Systems in
Metal Forming, Advanced Technology of Plasticity, 1999. v. 1. p. 117-128.
11. STEININGER, V.: Forming Simulation in the Vehicle Development Process, The
Fabricator (1998) p. 1-6.

106