Integrated behavior modeling of space-intensive mechatronic systems [Elektronische Ressource] / Benjamin Hummel

Informations

• Publié par: technische_universitat_munchen
• Publié le: 01 janvier 2011
• Nombre de lectures: 20
• Langue: English
• Poids de l'ouvrage: 10 Mo

Extrait

Integrated Behavior Modeling
of Space-Intensive Mechatronic Systems
BenjaminHummelInstitut fur¨ Informatik
der Technischen Universitat¨ Munchen¨
Integrated Behavior Modeling
of Space-Intensive Mechatronic Systems
BenjaminHummel
Vollstandiger¨ Abdruck der von der Fakultat¨ fur¨ Informatik der Technischen Universitat¨
Munchen¨ zur Erlangung des akademischen Grades eines
Doktors der Naturwissenschaften (Dr. rer. nat.)
genehmigten Dissertation.
Vorsitzender: Univ.-Prof. Dr. Johann Schlichter
Prufer¨ der Dissertation:
1. Univ.-Prof. Dr. Dr. h.c. Manfred Broy
2. Univ.-Prof. Dr. Wilhelm Schafer¨ ,
Universitat¨ Paderborn
Die Dissertation wurde am 25. August 2010 bei der Technischen Universitat¨ Munchen¨
eingereicht und durch die Fakultat¨ fur¨ Informatik am 25. Januar 2011 angenommen.ivAbstract
Many of today’s systems perform their tasks by a complex combination of electrical and mechanical
effects with programmable logic controllers. Such systems are commonly referred to as mechatronic
or cyber-physical systems. For a large class of such systems, the notion of space is at the very core
of their mode of operation, as they measure and affect the spatial relationship of physical objects.
Examples of such systems are commonly found in the domain of factory automation and machine
tools, where products are transported, manipulated, and assembled.
While there are many different models for capturing static and structural aspects of those systems in
mechanical engineering, and computer science provides an abundance of behavior models for pure
software systems, neither of these can describe and explain the operation of such systems when
used in isolation. Full understanding of such a system can only be achieved by using a model that
integrates the spatial and the behavioral view. Combined approaches exist, but are usually limited
to the connection of different models by structural links without clearly defined semantics.
This lack of suitable modeling techniques has a practical impact, as the increasing complexity of
systems built today pushes traditional development processes to their limits. In these processes the
disciplines of mechanics, electrics, and software usually act isolated from each other, which inhibits
taking advantage of the full potential of mechatronic solutions. One ingredient to overcome this
separation of engineering disciplines is an integrated abstract model that serves as a common lan-
guage for documentation and logic design of the system. By modeling and simulating the system on
an abstract level, design alternatives can be explored and a common interdisciplinary understanding
of the internal workings of the system can be fostered.
This thesis proposes an integrated model for space-intensive mechatronic systems that tightly cou-
ples both the structural spatial aspects and the temporal and logic behavior of a system. Compared to
other approaches, strong emphasis is on the semantic meaning and mathematical foundation of the
model. Such a modeling theory not only contributes to the academic discussion on suitable models
but also provides a solid basis for formal analysis and construction of modeling and engineering
tools.
To evaluate the adequateness of the model from an engineering perspective, this thesis also discusses
the operationalization of the theoretical model and describes means by which the gap to the practical
application can be closed. The actual tool prototype that implements the model is used in two case
studies with machine tool vendors to investigate to which extent the behavior of real-world systems
can be captured by the model.
vAbstract
viAcknowledgments
The result of a task that spans several years, such as this thesis, is never the success of a single
person, but rather depends on a multitude of supporters to whom I am deeply grateful. First of all, I
want to thank my supervisor Manfred Broy for the opportunity to work in his group and his support
over all the years. The impressive working environment at his chair and his ability to ask the right
questions at the right time contributed greatly to this thesis. I also want to thank Wilhelm Schafer¨
for accepting to be my co-supervisor and for the insightful and enjoyable discussion we had during
my visit in Paderborn.
I am grateful for my colleagues’ support over the years, for all I could learn from them, and the
many (sometimes heated) debates and discussions we had. I especially want to mention Jewgenij
Botaschanjan, Peter Braun, Florian Deißenbock,¨ Martin Feilkas, Alexander Harhurin, Lars Heine-
mann, Markus Herrmannsdorfer¨ , Florian Holzl,¨ Elmar Jur¨ gens, Birgit Penzenstadler, Silke Muller¨ ,
Bernhard Schatz,¨ Judith Thyssen, Stefan Wagner, and Sebastian Winter, who all impacted my work
in one or another way. It was and still is a pleasure to work with all of you! Special thanks go to
Florian Deißenbock,¨ who endured the tedious task of reading the largest part of my thesis, and to
Judith Thyssen, who read the core chapters (and is the only person I know, who not only reads all
formulas but also spots the mistakes in them).
Several students contributed to the tool prototype and its underlying framework, and I want to thank
Christoph Dobber¨ , Christoph Klaaßen, Daniela Steidl, and Andreas Wandinger for their work.
Furthermore, I thank Alexander Lindworsky from the engineering department for teaching me a lot
about mechanical engineering and the automation domain. We collaborated in multiple projects and
his different perspective often lead to new insights. In this context, I also want to thank the industry
partners of the AutoVIBN project for the many discussions we had during the milestone meetings.
Especially Jur¨ gen Franz and his critical questions helped me to better understand the subject matter
and the practitioner’s view.
For their ongoing support I want to thank my family and friends. They helped me to put everything
into perspective when stuck with my work and endured me even when being slightly off. My father
also deserves a big thank you for checking my spelling in the final version.
Thanks also go to the local Aikidoka at Ismaning for helping with the practical aspects of classical
mechanics; there is no better way to relax after work than approaching the mat at high velocity.
viiAcknowledgments
viiiContents
Abstract v
Acknowledgments vii
1 Introduction 1
1.1 Motivating Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.1 Space-Intensive Mechatronic Systems . . . . . . . . . . . . . . . . . . . . 4
1.2.2 Integrated Behavior Modeling . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3.1 Lack of Practical Models . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3.2 Lack of Formal . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 Contribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.5 Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2 Preliminary Considerations 11
2.1 The Role of Abstraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 The Role of Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Problems of Behavior Models with Space . . . . . . . . . . . . . . . . . . . . . . 14
2.4 Requirements to a Suitable Modeling Technique . . . . . . . . . . . . . . . . . . . 17
2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3 Related Work 21
3.1 Formal Models of Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2 Models for Hybrid Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.3 Integrated Meta-Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.4 Virtual Commissioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.5 Simulation-Centric Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.6 Mechatronic and Systems Engineering Models . . . . . . . . . . . . . . . . . . . 35
3.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4 Space and Time 39
4.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
ixContents
4.2 Formalizing Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.2.1 Common Notions of Space . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.2.2 Transformable Collision Space . . . . . . . . . . . . . . . . . . . . . . . . 43
4.2.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.3 Formalizing Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.3.1 Linear Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.3.2 Streams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.4 Relating Time and Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5 Modeling Spatio-Temporal Systems 55
5.1 Soft