Home » Proposal » Hilding elmqvist phd thesis proposal

Hilding elmqvist phd thesis proposal

Introduction

Dymola was created by Hilding Elmqvist from the Technical College at Lund (LTH) in Norway within his PhD dissertation.

Dymola is really a modeling atmosphere that allows the consumer to map hardware aspects of physical systems directly onto corresponding software components.

Elmqvist recognized there mustn’t exist any among input and output parameters of these objects. The direction of knowledge flow are only able to be determined throughout the selection of the models in line with the embedding from the model components in to the overall model.

Since that time, Dymola/Modelica matured in to the best and many flexible modeling atmosphere for that description of physical systems presently on the program market.

Historic Development

In 1978, Hilding Elmqvist introduced an initial form of Dymola within his PhD dissertation. The word what presented in those days offered a purely alphanumerical interface that today forms the foundation from the modern modeling language Modelica. An initial form of the Dymola compiler have been created in Simula 68. However, that version was soon substituted with a more recent version which was created in Pascal. The compiler converted Dymola models to Simnon. a simulation atmosphere that Elmqvist acquired in 1975 within his MS Thesis.

Almost 30 Years Ago, Hilding Elmqvist created a first form of a picture interface (Graphical user interface) that will let the user to explain an actual system by way of graphical objects. The graphic modules were then mapped onto Dymola. Elmqvist was believing that graphical descriptions of systems were simpler to handle, because they support two-dimensional connections, whereas an alphanumerical description must always be one-dimensional. However, the suggested Graphical user interface couldn’t meaningfully be applied, because it consumed to a lot of computer sources. The computers of individuals days weren’t yet effective enough to meaningfully support this type of development.

After Elmqvist had left academia back in 1984 to visualize employment in industry, I personally along with my students ongoing in 1986 with the introduction of Dymola. Particularly, we developed new translation modules that mapped Dymola models to the simulation languages Desire and ACSL. These modules were developed within grant from NASA.

In 1991, I printed it Continuous System Modeling [1], by which Dymola performed a main role. This book convinced Hilding Elmqvist to stop his position in industry and make rather a business of their own, known as Dynasim AS. the only reason for which will be the further growth and development of Dymola.

A still unsolved problem was the symbolic management of algebraic loops and structural singularities, because they happened frequently within the coupling of Dymola models. This issue was solved by Elmqvist in 1992. It brought to publication [2].

Within the summer time of 1992, I labored in the German Aerospace Establishment (DLR) in Oberpfaffenhofen. This contract brought to some close and incredibly fruitful collaboration between Hilding Elmqvist (Dynasim), Martin Otter (DLR), and myself. Within the follow up, DLR financed majorly the further growth and development of Dymola for quite some time. Like a first project, an interface towards the DLR simulation atmosphere DSblock was created that forms the foundation from the Dymosim simulation atmosphere being used today.

In 1993, the Dymola modeling atmosphere was enriched by language elements for modeling discontinuous processes that replaced and improved around the modeling concepts that were suggested within the modeling atmosphere COSY [3]. In parallel, a brand new form of the Dymola compiler was issued, now created in C.

In 1994, the symbolic management of algebraic loops was improved by the development of tearing variables. Also introduced were new language factors that should let the user to explain vector processes, because they occur for instance in mechanics within the description of multibody systems (MBS), by way of matrices, supplying a interface much like that obtainable in Matlab.

And in 1994, a brand new Graphical user interface was introduced that first required the Dymodraw. As opposed to the sooner growth and development of 1981, Dymodraw was completely functional. Regrettably, he tools offered in Dymodraw for the phrase new graphical objects remaining much to become preferred. Because of this, it had been simple to use already developed graphical objects in Dymodraw, but the style of new graphical objects was painful and error prone. The graphic compiler was created in C++.

In 1995, the efficiency from the compiled simulation code was considerably improved by the development of DASSL because the default integration method, in addition to through the new idea of “inlining”, i.e. the merging from the integrator equations using the model equations before the symbolic preprocessing from the resulting differential-algebraic equation (DAE) system [4].

In 1996, the symbolic preprocessor was enhanced by another formula, known as “relaxing” of straight line equation systems. The relaxing formula provides a symbolic implementation from the Gauss reduction formula [5].

And in 1996, the word what meaning of Dymola, which right now was 18 years of age, went through a significant revision. Die resulting new language was known as Modelica. Modelica’s language definition was put into the general public domain. Because the name “Dymola” have been liberated, the Graphical user interface in the future assumed the name Dymola. By 2002, Dymola Version 5 was launched, which finally offered a totally redesigned and completely functional Graphical user interface, by which even the development and design of recent graphical objects can be achieved easily and reliably.

In 1997, a typical committee was produced, the job which was the further development and standardization from the language meaning of Modelica. Regrettably, I possibly could no more participate personally within this effort, because this might have known as for just two-3 journeys to Europe yearly, mostly during teaching sessions, that we couldn’t justify.

Already in 1993, I ended using the further growth and development of the Dymola compiler, as I wouldn’t and may not contend with Elmqvist’s group within this arena. Rather, I subsequently focused my efforts on the task that my students and that i could handle a lot more easily and effectively, namely the development and design of recent Dymola/Modelica libraries, for example BondLib.

Most Significant Publications

Elmqvist, H. F.E. Cellier, and M. Otter (1993), Object-Oriented Modeling of Hybrid Systems. Proc. ESS’93, SCS European Simulation Symposium. Delft, Holland, pp.xxxi-xli.

Otter, M. H. Elmqvist, and F.E. Cellier (1996), “Relaxing” – A Symbolic Sparse Matrix Method Exploiting the Model Structure in Generating Efficient Simulation Code. Proc. Symposium on Modelling, Analysis, and Simulation. CESA’96, IMACS MultiConference on Computational Engineering in Systems Applications, Lille, France, vol.1, pp.1-12.