Conference PaperPDF Available

Using Static Parametric Design to Support Systems Engineering of Industrial Automation Systems

Authors:
  • Bosch Rexroth, Germany, Lohr am Main

Abstract and Figures

This paper proposes a static parametric design methodology for application of the model based systems engineering (MBSE) paradigm in the world of Modelica. This methodology allows for parameter synthesis of the industrial automation systems under consideration of customer requirements. Furthermore, the parametrized system can be verified automatically. An integrated system model consisting of requirements, system design and verification models is created and can be used as a design template to generate a new parameter set according to the change of customer requirements. A case study from the practice is presented to proof the concept of this methodology
Content may be subject to copyright.
Using Static Parametric Design to Support Systems Engineering
of Industrial Automation Systems
Hongchao Ji1Lars Mikelsons1Karl Kempf1Dieter Schramm2
1Bosch Rexroth AG, Lohr am Main, Germany
{hongchao.ji, lars.mikelsons, karl.kempf}@boschrexroth.de
2University of Duisburg-Essen, Duisburg, Germany
dieter.schramm@uni-due.de
Abstract
This paper proposes a static parametric design
methodology for application of the model based sys-
tems engineering (MBSE) paradigm in the world of
Modelica. This methodology allows for parameter
synthesis of the industrial automation systems under
consideration of customer requirements. Furthermore,
the parametrized system can be verified automatically.
An integrated system model consisting of require-
ments, system design and verification models is cre-
ated and can be used as a design template to generate
a new parameter set according to the change of cus-
tomer requirements. A case study from the practice is
presented to proof the concept of this methodology.
Keywords: Model Based Systems Engineering,
SysML, Modelica, Parameter Synthesis
1 Introduction
The complexity of modern industrial automation sys-
tems increases steadily. New functions and technolo-
gies need to be integrated to fulfill customer require-
ments, environmental regulations and/or safety stan-
dards. The increasing complexity has raised many
challenges such as keeping the the design consistent
and approving the correctness with respect to the cus-
tomer requirements. Model based systems engineer-
ing (MBSE) is defined as the formalized application
of modeling to support system requirements, design,
analysis, verification and validation activities begin-
ning in the conceptual design phase and continuing
throughout development and later life cycle phases.
Hence MBSE is a suitable approach to cope with these
challenges.
One of the key issue in MBSE process is to deter-
mine the proper dimension of the system design ac-
cording to the formalized requirements model. The
static parametric design methodology uses Modelica
static models together with the dynamic models to
support the MBSE process by the means of select-
ing the proper components of the desired system from
given product catalogs, dimensioning the sub-systems
as well as checking the correctness of the system de-
sign with respect to systems requirements.
The objective of the static parametric design
methodology is to perform a parameter synthesis of
a technical system according to the customer require-
ments automatically. Furthermore, the calculated sys-
tem design can be verified automatically as well. The
Systems Modelling Language (SysML) [11] is used to
formalize the customer requirements. Moreover, the
extension of abstraction levels and classification de-
fined in [4] is also applied in this paper. Work on
the integration of SysML and Modelica has already
proven its effectiveness in the MBSE [6,8,9]. Reusing
these improvements the SysML models can be trans-
formed into executable Modelica models.
In this contribution, we focus on the standard appli-
cation that the structure of the desired system is nor-
mally known to the system engineers. The require-
ments models serve as the basis of the static para-
metric design methodology. By changing the require-
ments models, a new parameter set of the system can
be obtained automatically. In this sense, the integrated
model consisting of requirements, system design and
verification models can be seen as a design template
for a standard application.
This paper is organized in 6 sections. Section 2il-
lustrates the current systems engineering process and
the need of model based systems engineering and
static parametric design methodology. As part of re-
lated work in Section 3a short introduction to SysML
and its integration with Modelica are given. Section
4introduces the static parametric design methodology
DOI Proceedings of the 9th International Modelica Conference 971
10.3384/ecp12076971 September 3-5, 2012, Munich, Germany

to support the systems engineering in detail. The capa-
bilities of the proposed methodology are demonstrated
using an industrial application in Section 5. The paper
closes with conclusions and an outlook to future work.
2 Systems Engineering of Industrial
Automation Systems
The systems engineering process is described in the
following referring to the well known V Model ac-
cording to the VDI 2206 standard [12] depicted in
(Figure 1).
Requirement Product
System Design
System Integration
Domain-specific Design
Mechanical Engineering
Electrical Engineering
Information Technology
Modelling and Model Analysis
Assurance of Properties
Figure 1: V Model According to VDI 2206 [12]
The main tasks of current systems engineering of
industrial automation systems can be summarized as
follows:
State the customer needs correctly and unam-
biguously;
Define the proper system design based on the cus-
tomer requirements;
Verify the system design against customer re-
quirements.
They will be introduced respectively in the following.
2.1 Requirements Specification
Due to the fact that the requirement specification is
the subject matter of contract between customer and
contractor the above described context implies that the
requirements engineering has to be seen not only from
the technical perspective but also needs to consider the
business process and the contractual situation along
the supplier chain. In this context it is self-evident that
the requirements shall be defined and structured not
only according to technical aspects but also according
to the contractual situation. The definition of levels of
abstraction is an appropriate way to meet these needs.
The depicted levels of abstraction in Figure 2reflect
the described supplier chain and major technologies
involved and therefore are a reasonable choice in the
context of automation systems. In order to deal with
the complexity of large systems the design objects are
clustered in a system break down structure. The re-
quirements derived on the different levels of abstrac-
tion can be referenced in requirement specifications in
order to provide the contractual views on subsets of
requirements.
Figure 2: Levels of Abstraction in Requirements and
System Design
2.2 Systems Design
Industrial automation systems are characterized by
their ability to process a material or work piece ac-
cording to a defined procedure to achieve the output of
a desired product. The challenge of the system engi-
neer is to design a machine that is capable to run the
process in a deterministic and efficient way. This task
is typically performed within a specific design domain
that refers to a field of technical expertise. The proper
selection of the components and their integration into
the overall structure strongly influence the function,
performance, robustness and reliability of the whole
system. Currently the selection of the components is
mainly determined by the competence of the system
engineers which is time consuming and error prone.
In order to avoid manual errors, it is desired to select
the components in a systematical manner.
Today, parameter synthesis of a technical system is
usually based on static calculations in the field of in-
dustrial automation systems. A small example of such
static calculation for selection of a hydraulic valve in
a hydraulic lift system is depicted in Figure 3. The di-
mension of the cylinder shall be first defined in order
to calculate the maximal flow rate through the valve
Using Static Parametric Design to Support Systems Engineering of Industrial Automation …
972 Proceedings of the 9th International Modelica Conference DOI
September 3-5, 2012, Munich Germany 10.3384/ecp12076971

with the given maximal cylinder velocity by
Qmax =AD·vmax.(1)
After that the nominal size of the valve is determined
by some design criteria. Moreover, the design crite-
ria origin from customer requirements as well. In this
case, the nominal size is defined by the nominal flow
rate which is calculated by
Qnominal =1.5·Qmax.(2)
The nominal size of the valve can be chosen from the
product catalog according to the nominal flow rate.
That means a proper component, in this case the valve,
is correctly selected for the desired system.
Figure 3: Schematic of a Hydraulic Lift System
In order to ensure the acceptance of a MBSE tool
offering the methodology presented in this paper, these
design guidelines have to be integrated into the MBSE
tool.
2.3 Verification and Validation
Model based verification and validation of a systems
design against systems requirements has been widely
used in the field of industrial automation systems. The
goal here is to verify the system design in an auto-
mated and reproducible way. Since this issue has al-
ready been addressed in virtual verification of systems
design against systems requirements (vVDR) method-
ology [10], it is used to approve the systems design in
this paper as well.
2.4 The Challenges
Since industrial automation systems usually consist of
components from different domains, it is hard to keep
the design correct and consistent. In order to deal with
this fact, the systems design from different engineering
domains shall be integrated into the whole MBSE pro-
cess. Therefore, a universal and standardized model-
ing language is required which shares the understand-
ing among engineers from different disciplines. This
common language shall enable the generation of re-
quirements models, system design models, traceabil-
ity models as well as verification models containing
domain-specific details. SysML is being proposed to
meet this requirement. However it has been evaluated
as not sufficient due to the lack of executable seman-
tics. Integration of the languages SysML and Mod-
elica has proven its efficiency in the area of MBSE
[5,6,7,10]. Therefore, in this contribution SysML
and Modelica are chosen as the modelling languages
applied in our MBSE process as well.
In order to set up a MBSE tool for parameter syn-
thesis, the following two questions have to answered:
1. How to perform a parameter synthesis to deter-
mine all the proper components based on the cus-
tomer requirements in an automatic manner?
2. How to link different kinds of models in the
whole MBSE process?
These two challenges have been addressed in this
paper by the means of using static parametric design
methodology in an integrated system model based on
SysML and Modelica. The details are presented in
Section 4.
3 Background and Related Work
3.1 The SysML and Modelica
SysML is a general purpose language used in the field
of systems engineering. It is defined as a UML pro-
file which reuses subsets of UML constructs and ex-
tends them with some additional modeling elements.
SysML is capable to capture the textual requirements
and to allocate them with the design models and test
cases. However due to loosely defined executable se-
mantics SysML is not capable to execute the modeled
physical systems. In contrast to that, Modelica is an
object oriented and equation based modeling language
for multi-domain physical systems. Graphical model-
ing is supported by the object diagram which offers an
Hongchao Ji, Lars Mikelsons, Karl Kempf and Dieter Schramm
DOI Proceedings of the 9th International Modelica Conference 973
10.3384/ecp12076971 September 3-5, 2012, Munich, Germany
intuitive way to describe power transmission through
acausal connections as well as directed signal flows.
Strong semantics allow the generation of executable
models of continuous as well as discrete systems. Ob-
ject oriented language constructs enable the efficient
reuse of models and the design of comprehensive and
easy to use model libraries. As mentioned in Section 2,
a language which integrates the descriptive modeling
power of SysML and the formal executable simulation
power of Modelica seems to be a promising approach
for the systems engineering in industrial automation
systems.
3.2 Related Work
Several work has already been done towards appli-
cation of MBSE paradigm using Modelica language
with different concerns. The vVDR methodology [10]
addresses mainly the virtual verification of systems
requirements by using UML, Modelica and Modeli-
caML. In Dubois et. al. [2] a requirement traceabil-
ity model to enforce the traceability concept in SysML
in the automotive domain is presented. Requirements
management and allocation have already been covered
in the other paper of the author [4].
This paper describes a methodology for the param-
eter synthesis of technical systems according to cus-
tomer requirements. Moreover, the different kinds of
models are linked with the other models in the inte-
grated system model and therefore it is easy to regen-
erate and to verify the final parametrized system. They
will be introduced in detail in the following section.
4 Static Parametric Design
The proposed static parametric design methodology is
based on Modelica static models. Static models are
defined as models that are constant over time. Consid-
ering the whole MBSE process, the following items
can be formalized as Modelica static models:
Requirement Specifications,
Product Catalogs,
Selection Criteria.
4.1 Definitions of Models
The requirement specifications can be defined as re-
quirements models which are captured as stereotyped
SysML models according to the different abstraction
levels and classifications in [4].
This classification is mainly based on the taxonomy
proposed in [3] with some changes as presented in
the following. Instead of the requirement type spe-
cific quality, the structural requirement is defined in
the field of industrial automation.
Afunctional requirement is the requirement that
should produce an expected reaction to a given
stimuli.
Aperformance is the requirement to check
whether a system variable such as timing, speed,
volume or throughput is in a desired range.
Astructural requirement is the requirement
which describes the structural demand of the
stakeholder.
Aconstraint is the requirement to provide the
technical and safety boundary conditions that the
system shall satisfy.
The product catalogs models are easy to understand
as Modelica static models. They can modeled as a
simple record class with some table definitions. In
this work, a UML library with a sub-set of the Bosch
Rexroth catalog is implemented and later transformed
to Modelica static models. The advantages of imple-
mentation as UML library over Modelica library is the
compatibility with SysML and the extendibility of the
product catalog.
The selection criteria are implemented as static cal-
culation models. The idea is using Modelica functions
to determine the proper dimension of the components.
Normally, those selection criteria for the components
are usually the same. Due to the fact of reuseability, a
Modelica library called ParametricDesign which con-
sists of most selection criteria in the field of industrial
automation systems is built as shown in Figure 4.
Besides the static models, simulation model is an
executable model which is used for dynamic simula-
tion. In the verification and validation phase, the sim-
ulation model is linked with the test cases to check the
correctness of the parametrized system design.
All those models are further transformed into Mod-
elica static models with the help of a Modelica code
generator, which is implemented with the help of
Eclipse Acceleo [1].
4.2 Link of Different Models
Several models have been defined in this static para-
metric design methodology. It is necessary to link
Using Static Parametric Design to Support Systems Engineering of Industrial Automation …
974 Proceedings of the 9th International Modelica Conference DOI
September 3-5, 2012, Munich Germany 10.3384/ecp12076971
Figure 4: Structure of the Parametric Design Library
those models in a efficient manner in order to per-
form a parameter synthesis automatically. The basic
idea is to reference the attributes of SysML model to
the variables, parameters and constants of the Mod-
elica model. Currently, these relations have been es-
tablished manually which is time consuming and error
prone. A method to extend the standard relationships
such as «satisfy», «verify» and «derive» has been pro-
posed in [4]. An overview of the linking of different
models in this methodology is shown in Figure 5. The
tooling that supports the binding of related objects is
implemented in Eclipse.
Figure 5: Link of Different Models
4.3 Methodology Description
The prerequests of application of the static design
methodology are
hydraulic library,
static design library,
product catalog library.
Furthermore, the formalized requirements models and
at least one simulation model have to be created at first
as the basis for application of the static parametric de-
sign methodology.
The main steps of this methodology can be summa-
rized as follows:
1. Capture the customer requirements as stereo-
typed requirements according to the proposed
classification in [4].
2. Create a simulation model from the hydraulic li-
brary on the considered level of abstraction.
3. Select the proper design criteria from the para-
metric design library and create the static calcu-
lation model.
4. Link the requirements model, static calculation
model as well as product catalog model in the
parametric design model.
5. Run a parameter synthesis to obtain a best suited
parameter set and the other possible parameter
sets for the desired system.
6. Set the obtained possible parameter sets in the
simulation model and save them as design vari-
ants.
7. Define test cases that need to satisfy customer re-
quirements.
8. Link the requirements model, test cases as well
as simulation models in the verification model.
9. Run a verification that executes all related test
cases and design variants.
10. Choose the best suited design variants according
to the verification results.
Hongchao Ji, Lars Mikelsons, Karl Kempf and Dieter Schramm
DOI Proceedings of the 9th International Modelica Conference 975
10.3384/ecp12076971 September 3-5, 2012, Munich, Germany
5 Application Example
In this section, a hydraulic lift system is used to
demonstrate the static parametric design methodology.
The hydraulic lift system is used to lift a load to a given
height. It shall be considered in the context of the
OEM-supplier relation as it applies to a typical Bosch
Rexroth engineering project.
The task of this case study is to define a best suited
parameter set of the desired lift system which ful-
fills all the customers requirements as well as techni-
cal constraints. First of all, a simulation model shall
be created. Therefore, the structure of the hydraulic
lift system has to be known for the system engineers.
Then, the static calculation model shall be created as
well by selecting the proper design criteria from the
parametric design library. After linking of different
models in the integrated system model, a parameter
synthesis can be performed to obtain the best combi-
nation of the components with the minimal dimension
which satisfy all the requirements.
The main advantage is that the system engineers can
use the integrated system model as a design template.
With the help of this design template, it is much more
easier to variate the parameter set of the hydraulic lift
system by changing the customer requirements auto-
matically.
5.1 Requirements Capture
The requirements from the customers are formalized
as follows: a load of 3000 kg shall be lifted to 0.5m
within 2 s. Besides the customer requirement there are
some technical constraints of the desired system. For
example, the maximum velocity can not exceed 0.6
m/s. The other important constraint is that the pres-
sure drop over the proportional valve shall not exceed
30% of the working pressure. The important require-
ments are listed in Table 1. The beginning letter of
the ID of the requirement refers to the type of the re-
quirement. The requirements P1, C2 and C3 can be
verified by test cases which are modeled as Modelica
models. Since the structral requirments S4 and S5 pro-
vide only the important design parameters, they are not
necessary or possible to verify. According to those re-
quirements, the proper components from the product
catalogs shall be selected iteratively until all the com-
ponents are chosen. They can be formalized as SysML
requirements model and later transformed into Mod-
elica static model. An example of the requirements
model and its generated Modelica code are shown in
Figure 6.
ID Description
P1 The load shall be lifted to 0.5mwithin 2 s.
C2 The max. velocity shall not exceed 0.6m/s.
C3 The pressure loss over the valve shall not
exceed 30% of the working pressure.
S4 The mass of the load is 3000 kg.
S5 The working pressure is 200 bar.
Table 1: The System Requirements List
Figure 6: An Example of Requirements Model and
Generated Modelica Code
5.2 Modelling of the Hydraulic Lift System
The object diagram (Figure 7) shows the structure of
the hydraulic lift system. The load is lifted by a differ-
ential cylinder which is driven by a constant pressure
source. The proportional valve is controlled by a sim-
ple P-controller to realize the position control. Since
the focus of this work is to illustrate the static paramet-
ric design methodology, the details about the model
will not be introduced here.
5.3 Static Parametric Design Process
The involved components from the product catalog
are the differential cylinder and proportional valve.
Hence, the design criteria models for those two com-
ponents in the design library are selected into the static
parametric design model together with the require-
ments model. Table 2and 3show the important de-
sign variables from the product catalogs of differential
cylinder and proportional valve.
Using Static Parametric Design to Support Systems Engineering of Industrial Automation …
976 Proceedings of the 9th International Modelica Conference DOI
September 3-5, 2012, Munich Germany 10.3384/ecp12076971
Figure 7: Object Diagram of the Hydraulic Lift Sys-
tem
Piston Rod Max
Diameter Diameter Stroke
mm mm mm
40 28 2000
50 36 2000
63 45 2000
80 56 2000
100 70 3000
125 90 3000
140 100 3000
Table 2: Product Catalog of Differential Cylinders
Nominal Max Pressure
Size Flow Rate Drop
l/min bar
10 170 80
16 450 180
25 900 350
27 1000 430
35 3500 1100
Table 3: Product Catalog of Proportional Valves
The requirements variables defined in the require-
ments models, such as mass of load, maximum veloc-
ity and desired lifting position are taken as inputs for
the design criteria which are implemented as Modelica
functions. Finally, a parametric design model is ob-
tained by linking the requirements models, static cal-
culation models and the related product catalog mod-
els.
After the parametric design model is created, the
automatic parameter synthesis can be done very con-
veniently. The parametric design model is interactive
solved and the design variables are calculated. Ac-
cording to those design variables the corresponding
components with the proper size are chosen until all
the components from the product catalog are chosen.
The following table shows the automatic generated
best suite combination of cylinder and valve, which
is defined as an optimal design. It is worth to mention
Piston Rod Max
Cylinder Diameter Diameter Stroke
mm mm mm
Optimal 100 70 3000
Nominal Max Pressure
Valve Size Flow Rate Drop
l/min bar
Optimal 25 900 350
Table 4: Selected Components from Product Catalog
that the gain of the P-controller is determined by "Try
and Error". In the future, this kind of parameter which
is not related to the product catalog can be defined by
the means of optimization.
5.4 Verification of System Design
After the static parametric design process is done, a
best suited set of combination of the components is
obtained. An automatic verification will check the
optimal design against customer requirements. This
is done by using vVDR methodology [10] to model
the test case of requirements with violation monitor.
According to the requirements definitions in Table 1,
three test cases are defined to verify the requirements
P1, C2 and C3. Figure 8shows the verification result
of this proposed optimal design. The first two figures
(Figure 8(a) and 8(b)) illustrate that the load is lifted to
0.5 meter after 2 second and does not exceed the max-
imal velocity. As shown in Fiugre 8(c), the pressure
loss over the valve in steady state satisfies the critical
30% of working pressure 200 bar as well. Therefore,
the hydraulic lift system with the automatic selected
parameter fulfills the customer requirement and tech-
nical constraints.
5.5 Comparison of Design Variants
The best suited combination of components from the
product catalog are supposed to have the minimal size
which satisfy all the requirements. It has been verified
to fulfill all the requirements in the last section. Nev-
ertheless, it is still not proved that the performance of
this design is better than the others variants. In order
Hongchao Ji, Lars Mikelsons, Karl Kempf and Dieter Schramm
DOI Proceedings of the 9th International Modelica Conference 977
10.3384/ecp12076971 September 3-5, 2012, Munich, Germany
0 0.5 1 1.5 2 2.5 3
0
0.2
0.4
t[s]
s[m]
Load Position
(a) Load Position
0 0.5 1 1.5 2 2.5 3
−0.5
0
0.5
1
t[s]
v[m/s]
Load Velocity
(b) Load Velocity
0 0.5 1 1.5 2 2.5 3
0
50
100
150
200
t[s]
p[bar]
Pressure Loss
(c) Pressure Loss
Figure 8: Verification Results of Optimal Design
to validate this, design variants around the optimal de-
sign with other nominal sizes can be generated from
this methodology. The design variants are defined by
substituting the components of the optimal design with
a smaller or larger nominal size. In this case, four de-
sign variants are automatically generated and used to
compare with the optimal design. The dimension of all
the design variants are shown in the following table.
Design Piston Rod Valve
Variants Diameter Diameter Size
Optimal 100 70 25
Variant 1 80 56 25
Variant 2 100 70 16
Variant 3 100 70 27
Variant 4 125 90 25
Table 5: Dimensions and Costs of Design Variants
The simulation results of the optimal design and the
other four design variants are shown in Figure 9. It
shows that the design variant 1 and 3 can approach the
desired position within 2 seconds. However the veloc-
ities exceed the maximal velocity constraint 0.6m/s.
The design variant 2 and 4 fulfill the second test case
and can not satisfy the first one. The results concern-
ing the third test case are listed in Table 6.
A verification matrix of the design variants against
0 0.5 1 1.5 2 2.5 3
0
0.1
0.2
0.3
0.4
0.5
Comparison Load Position
Desired
Optimal
Variant 1
Variant 2
Variant 3
Variant 4
t[s]
s[m]
(a) Comparison of Load Position
0 0.5 1 1.5 2 2.5 3
−0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4 Comparison Load Velocity
Maximal
Optimal
Variant 1
Variant 2
Variant 3
Variant 4
t[s]
s[m]
(b) Comparison of Load Velocity
Figure 9: Comparison of Cylinder Positions and Ve-
locities of Different Design Variants
the test cases are shown in Table 6. Furthermore, the
costs of design variants depending on the dimensions
of selected components from the product catalog can
also be calculated. With the help of this verification
matrix and the price, the selected optimal combination
of the components from the product catalog is proved
to be exact the best suited design.
Design Test Test Test System
Variants Case 1 Case 2 Case 3 Cost
Optimal passed passed passed 2000
Variant 1 passed failed passed 1700
Variant 2 failed passed passed 1820
Variant 3 passed failed failed 2040
Variant 4 failed passed failed 2400
Table 6: Verification Matrix and System Costs
Using Static Parametric Design to Support Systems Engineering of Industrial Automation …
978 Proceedings of the 9th International Modelica Conference DOI
September 3-5, 2012, Munich Germany 10.3384/ecp12076971
5.6 Open Issues
This case study demonstrates the proposed static pa-
rameter design methodology. According to the cus-
tomer requirements and technical constraints, the di-
mension of the desired system can be defined auto-
matically. However, the main drawback is that the
simulation model shall be first modeled. That means
this methodology can not be applied to arbitrary sys-
tem. This is due to the fact that there is not enough
information for determining a proper combination of
the desired system in the practice. This drawback also
limit the application of MBSE in the field of industrial
automation systems.
It is noticed that the order for the selection of com-
ponents is fixed in this case, i.e., the dimension of
cylinder shall be first defined in order to determine the
nominal size of the proportional valve. Sometimes the
order is not fixed. For both cases the system engineers
shall have the chance to determine the order for the
selection of components more freely without reimple-
mentation of static design model. Since Modelica is a
standardize equation-based modelling language, it has
been chosen to meet these requirements. It is capable
to deal with the this issue. One proposed concept is
to switch the variability of parameter and variable of
static models in an arbitrary manner. The system en-
gineers can give the known parameters until the static
model is balanced and solvable to calculate the other
unknown variables.
6 Conclusion and Future Work
In this paper a static parametric design methodology
has been analyzed in the systems engineering con-
text of industrial automation systems. A set of pos-
sible design variants with different dimensions can be
automatically generated and compared by using this
methodology. The concept has been demonstrated by
a case study of a typical engineering project. The other
contribution of this work is allocation of this method-
ology in a MBSE process in which the parametrized
design variants are fully traceable to the other models.
In the future, the proposed methodology will be im-
plemented as an Eclipse plug-in for better tool sup-
port of the static parametric design. It is usually the
case, not all the parameters can be defined by the static
parametric design methodology. Integration of an op-
timizer to define those parameters is desired. Applica-
tion of a big scenario is also a part of future work.
Acknowledgments
This work is funded by Bosch Rexroth AG and Ger-
man Federal Ministry of Education and Research
(BMBF) in the ITEA2 OPENPROD project.
References
[1] http://www.acceleo.org.
[2] Hubert Dubois, Marie-Agnès Peraldi-Frati, and
Fadoi Lakhal. A Model for Requirements Trace-
ability in a Heterogeneous Model-Based Design
Process: Application to Automotive Embedded
Systems. In 15th IEEE International Conference
on Engineering of Complex Computer Systems,
ICECCS 2010, Oxford, United Kingdom, 22-26
March 2010, pages 233–242, 2010.
[3] Martin Glinz. On Non-Functional Requirements.
In 15th IEEE International Requirements Engi-
neering Conference, RE 2007, October 15-19th,
2007, New Delhi, India, pages 21–26, 2007.
[4] Hongchao Ji, Oliver Lenord, and Dieter
Schramm. A Model Driven Approach for
Requirements Engineering of Industrial Au-
tomation Systems. In Proceedings of the
4th International Workshop on Equation-
Based Object-Oriented Languages and Tools
(EOOLT’07), pages 13–24, 2011.
[5] Thomas Johnson, Christiaan J.J. Paredis, and
Roger Burkhart. Integrating Models and Simu-
lations of Continuous Dynamics into SysML. In
Proceedings of 6th International Modelica Con-
ference, Bielefeld, Germany, 3-4,March, 2008,
2008.
[6] Christiaan J.J. Paredis, Yves Bernard, Roger M.
Burkhart, Hans-Peter de Koning, Sanford
Friedenthal, Peter Fritzson, Nicolas F. Rou-
quette, and Wladimir Schamai. An Overview of
the SysML-Modelica Transformation Specifica-
tion. In 2010 INCOSE International Symposium,
July 2010.
[7] Adrian Pop, David Akhvlediani, and Peter Fritz-
son. Towards Unified System Modeling with
the ModelicaML UML Profile. In Proceedings
of the 1st International Workshop on Equation-
Based Object-Oriented Languages and Tools
(EOOLT’07), pages 13–24, 2007.
Hongchao Ji, Lars Mikelsons, Karl Kempf and Dieter Schramm
DOI Proceedings of the 9th International Modelica Conference 979
10.3384/ecp12076971 September 3-5, 2012, Munich, Germany
[8] Wladimir Schamai. Modelica Modeling Lan-
guage (ModelicaML). Technical report, EADS
Innovation Works, Germany, 2009.
[9] Wladimir Schamai, Peter Fritzson, Chris Pare-
dis, and Adrian Pop. Towards Unified Sys-
tem Modeling and Simulation with Modeli-
caML: Modeling of Executable Behavior Us-
ing Graphical Notations. In Proceedings of the
7th International Modelica Conference, Como,
Italy, 20-22 September 2009, number 43 in
Linköping Electronic Conference Proceedings,
pages 612–621. Linköping University Electronic
Press, Linköpings universitet, December 2009.
[10] Wladimir Schamai, Philipp Helle, Peter Fritz-
son, and Christiaan J. J. Paredis. Virtual Veri-
fication of System Designs against System Re-
quirements. In Models in Software Engineering
- Workshops and Symposia at MODELS 2010,
Oslo, Norway, October 2-8, 2010, Reports and
Revised Selected Papers, pages 75–89, 2010.
[11] http://www.omgsysml.org.
[12] VDI. Design Methodology for Mechatronic Sys-
tems (VDI 2206). Technical report, VDI, 2004.
Using Static Parametric Design to Support Systems Engineering of Industrial Automation …
980 Proceedings of the 9th International Modelica Conference DOI
September 3-5, 2012, Munich Germany 10.3384/ecp12076971
ResearchGate has not been able to resolve any citations for this publication.
Conference Paper
Full-text available
In order to support the development of complex products, modeling tools and processes need to support co-design of software and hardware in an integrated way. Modelica is the major object-oriented mathematical modeling language for component-oriented modeling of complex physical systems and UML is the dominant graphical modeling notation for software. In this paper we propose ModelicaML UML profile as an integration of Modelica and UML. The profile combines the major UML diagrams with Modelica graphic connec- tion diagrams and is based on the System Modeling Language (SysML) profile.
Conference Paper
Full-text available
This paper provides an overview of the formal trans formation between the two complementary languages: OMG SysML TM and Modelica. SysML is a standardized general pur pose graphical modeling language for capturing complex system descriptions in terms of their structure, behavior, properties, and requirements. Modelica is a standa rdized general purpose systems modeling language for analyzing the continuous and discrete time dynamics of complex systems in terms of differential algebraic equations. Integrating the descriptive power of SysML models with the analytic and computational power of Modelica models provides a capability that is significantly greater than provided by SysML or Modelica individually. A standardized bi-directional transformation between the two modeling languages is being developed that will support implementations to transfer efficiently and automat ically the modeling information between SysML and Modelica models without ambiguity. In addition to an overview of this bi-directional transformation approach, the paper provides a simpl e example to clarify the transformation principles and to illustrate the important synergie s resulting from the integration between these two languages.
Article
Full-text available
Model driven requirements engineering (MDRE) is proposed to deal with the ever-increasing complexity of technical systems in the sense of providing requirement specifications as formal models that are correct, complete, consistent, unambiguous and easy to read and easy to maintain. A critical issue in this area is the lack of a universal and standardized modeling language which covers the whole requirements engineering process from requirement specification, allocation to verification. SysML is being proposed to meet these requirements. In this paper a model driven requirements engineering process for industrial applications in the field of automation systems is described in order to reveal shortcomings in recent modeling tools and modeling languages. Special focus is layed on the requirements definition and the automated verification of the design against the requirements using executable models. Based on the analysis a new profile of the Unified Modeling Language (UML) called Model Driven Requirements Engineering for Bosch Rexroth (MDRE4BR) is presented which aims to contribute to latest investigations in this field. An application example of a hydrostatic press system is given to illustrate the approach.
Conference Paper
Full-text available
This paper is a further step towards application of the Model-Based Systems Engineering (MBSE) paradigm, using standardized, graphical, and executable system modeling languages. It presents further development of Modelica graphical Modeling Language (Modeli-caML), a UML Profile for Modelica, which enables an integrated modeling and simulation of system require-ments and design (for systems including both hardware and software). This approach combines the power of the OMG UML/SysML standardized graphical notation for system and software modeling, and the modeling and simulation power of Modelica. It facilitates the creation of executable system-specification and analy-sis models that can simulate time-discrete (or event-based) and time-continuous system behavior.
Article
Full-text available
In this paper, we combine modeling constructs from SysML and Modelica to improve the support for Model-Based Systems Engineering (MBSE). The Object Management Group has recently devel-oped the Systems Modeling Language (OMG SysML™). This visual modeling language provides a comprehensive set of diagrams and constructs for modeling many common aspects of systems engi-neering problems, such as system requirements, structures, functions, and behaviors. Complementing these SysML constructs, the Modelica language has emerged as a standard for modeling the continuous dynamics of systems in terms of hybrid discrete-event and differential algebraic equation systems. In this paper, the synergy between SysML and Mode-lica is explored at three different levels: the defini-tion of continuous dynamics models in SysML; the use of a triple graph grammar to maintain a bi-directional mapping between these SysML con-structs and the corresponding Modelica models; and the integration of simulation experiments with other SysML constructs to support MBSE. Throughout the paper, an example of a car suspension is used to demonstrate these contributions.
Conference Paper
Full-text available
Requirements traceability modeling is a key issue in real-time embedded design process. In such systems, requirements are of different nature (software-related, system-related, functional or non functional) and must be traced through a multilevel design flow which integrates multiple tools and heterogeneous models. Validation and Verification (V&V) activities must be performed on models and on the final product to check whether they match the initial requirements. Results of design and V&V activities must be able to impact traceability information. We thus propose DARWIN4REQ, a metamodel for requirement traceability, based on three independent flows (requirement model, solution model and V&V model). The new metamodel establishes a link between these flows and affords full traceability of requirements, including those set for heterogeneous models. This paper presents the DARWIN4REQ metamodel and its use in the context of heterogeneous models for requirements modeling, design and V&V. An automotive application illustrates the proposed approach based on UML-profiles such that SYSML, EAST-ADL2 and MARTE for design and on SIMULINK, SyNDEx and TIMESQUARE for V&V activities.
Conference Paper
Full-text available
System development and integration with a sufficient maturity at entry into service is a competitive challenge in the aerospace sector. With the ever-increasing complexity of products, this can only be achieved using efficient model-based techniques for system design as well as for system testing. However, natural language requirements engineering is an established technique that cannot be completely replaced for a number of reasons. This is a fact that has to be considered by any new approach. Building on the general idea of model-based systems engineering, we aim at building an integrated virtual verification environment for modeling systems, requirements, and test cases, so that system designs can be simulated and verified against the requirements in the early stages of system development. This paper provides a description of the virtual verification of system designs against system requirements methodology and exemplifies its application in a ModelicaML modeling environment.
Conference Paper
Although the term 'non-functional requirement' has been in use for more than 20 years, there is still no consensus in the requirements engineering community what non-functional requirements are and how we should elicit, document, and validate them. On the other hand, there is a unanimous consensus that non-functional requirements are important and can be critical for the success of a project. This paper surveys the existing definitions of the term, highlights and discusses the problems with the current definitions, and contributes concepts for overcoming these problems.
Methodology for Mechatronic Systems (VDI 2206)
  • Vdi
  • Design
VDI. Design Methodology for Mechatronic Systems (VDI 2206). Technical report, VDI, 2004.