Research Interests (under construction)
Multi-Formalism Modelling and Simulation is at the core
of my research.
The basis is the realisation that we study complex systems (and
their sub-systems) at different levels of abstraction. More
formally, different researchers will use different formalisms
(often determined by their background or application domain) such
as System Dynamics, Partial Differential Equations, Bond Graphs,
Petri Nets, Finite State Automata, Markov Chains, ... to model
(sub-)system behaviour. The challenge is to relate those
formalisms and to investigate how they can be meaningfully combined.
The research entails the theoretical background as well as the
implementation of software tools for studying complex systems.
The following article gives a brief
introduction to the Multi-Formalism concepts.
Research Projects (under construction)
To some (often minimal) degree, I am involved in all
BIOMATH projects.
Below are the projects which I am directly responsible for.
Basically, these are the projects of the
``deductive''
research group of the
BIOMATH department.
- FORmalisation of the Modelling and Simulation process (FORMS).
RUG Bijzonder Onderzoeks Fonds (BOF 01108194).
The goal of this project is to
investigate formal approaches to complex heterogeneous systems modelling.
Both abstract representations and models of the modelling process are
developed. A small case study demonstrates the viability of the approach.
Keywords: formal description, computer aided modelling, simulation,
process programming. This project is nearing completion.
The final report will be
available here. The
proposal for a
sequel project, MultiForms (in Dutch, gzip'd PostScript) is
only available locallly (from hobbes.rug.ac.be).
- Generic Interactive System for MOdelling and Simulation (GISMOS).
With Delft University of Technology, The Netherlands,
Cracow University of Technology,
Poland, and Riga Technical University, Latvia.
Mainly focus on Discrete Event models and the optimisation thereof.
-
Simulation in Europe Basic Research
Working Group (SiE-WG): coordination and actions related to
methodology, model semantics, and model specification languages.
- WEST++:
a modelling and simulation environment for design and optimisation
of Activated Sludge Waste Water Treatment Plants (WWTPs).
Features heuristics for
automated model choosing, efficient code generation for DAE simulations,
parameter estimation and optimisation, and a first attempt at automated
transformation of PDE's to DAE's (gzip'd
PostScript report).
With Aquafin
(Flemish Authority for Water Purification Infrastructure),
HEMMIS
(commercialisation and support, in particular, the dedicated
Windows95/NT version of the environment),
and EPAS (a WWTP consulting company).
The precursor of this effort was the WEST project.
-
CERES-Global Knowledge Network.
In Europe, the
CERES-EXP project (BIOMATH - Delft University of Technology)
develops an agent-based system for meaningful exchange of experiment
(real or simulated) information.
- Software Process Modelling (modelling of the software development
process) and Performance Modelling using stochastic models
and simulation.
- Symbolic techniques for Control Engineering: automatic stability
analysis and controller design for continuous systems
specified in state space as well as transfer function form.
A report of the PSYCO project
(gzip'd PostScript) gives an in-depth description. This prototype
was implemented in the MuPAD
computer algebra environment and in the above mentioned WEST++.
- The Modelica design group
of which I am a member is developing a "de facto" standard
modelling language for physical systems modelling. Modelica
is more pragmatic than our own MSL (Model Specification Language).
Note to students: if you are interested in doing a PhD or postDoc on one
of the above subjects, or on some related aspect of Modelling and Simulation
(and you expect to have sufficient grades etc. to
apply for a research grant),
contact me.
List of Activities (under construction)
- Under Construction:
Lecture notes on Modelling and
Simulation
Missing: Harel Charts, Petri Nets, Random Number Generation,
Statistical Analysis of Simulation Results, semantics of
Process Interaction Modelling Languages, Neural Networks
for Classification, Genetic Algorithms for Optimisation,
... (not yet in electronical form).
- On Wednesday, November 25, Dr. Suri Venkatachalam will
give a
lecture on "Organization of Biological Diversity".
His talk forms the basis for further collaboration on
behavioural modelling using System Dynamics and Cellular
Automata.
- Since 1997, I am an Associate Editor of the
SCS
Transactions on Simulation. Here are some
Instructions
for Referees.
- I spent the month February 1997 at the
Supercomputing Education
and Research Centre (SERC) of the
Indian Institute of Science (IISc) in Bangalore, India.
Apart from work on a theoretical basis for multi-formalism modelling
and simulation, I gave a few lectures on modelling and simulation.
My most "heroic" achievement was a one day intensive course on
that subject.
The online lecture notes have been superseded by those above
(1998).
The following announcement
describes the content of the course in slightly more detail.
Below are (most of) the transparencies I used for the course.
They are mostly figures and currently NOT
annotated. Still, many of the figures are self-explanatory.
The main aim of the talk is to show how different formalisms
and applications share some "generic" properties.
- July-October, 1996 I was a Research Associate at the
Concurrent Engineering Research Center
(CERC) in Morgantown, WV (USA). Here is a
short description of my work on SimFormatics at CERC.
Recent Publications (under construction)
