An object-oriented framework for modeling dynamic connectivity
Year: 1993
Type: dissertation
Abstract: Science and engineering rely on resolving systems into constituent components whenever possible to simplify the analysis and design process. This divide and conquer approach is useful in many of the problems commonly encountered. A system can be divided into components if it can be assumed that: points, edges, surfaces, and regions can be identified where connectivity relations can be reasonably established; and the constitutive equations within each of these interconnected components can be expressed in a localized manner involving only the variables of interconnections and other local variables. The resolution of systems into components is not only a common basis of the design processes that are used, but is at the heart of effective computer-based modeling and simulation.
The advancement to the state of computer-based modeling and simulation of systems presented in this dissertation deals with systems comprised of components whose characteristics, existence, and interconnections vary over time. A simulation framework based on recent results in object-oriented design is proposed. The framework defines a set of abstractions that serve as building blocks for describing connections, components and aggregations of components. The framework incorporates a set of concurrent objects which cooperatively respond to changes in the model topology and component population during model execution.
The construction of hierarchical models consisting of both continuous and discrete components is supported. The behavior of a component is represented by mathematical equations, and/or by blocks of code that are executed when a component sends or receives messages. Equations representing the connections among components are formulated by the framework when new connections occur and are revoked when the connections are removed. The equations representing connections and the equations representing component behavior are organized into systems of equations and solved on a demand basis.
DOI: 10.25335/4pgf-0j46
Author: Engelsma, Jonathan Robert
Institution: Michigan State University
Language: English
Subfield: Modeling and Simulation
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2464329773
