Abstract
Purpose: The main objective is to accelerate the mathematical modeling of complex systems and offer the researchers an accessible and standardized platform for model sharing and reusing. Methods: We describe a methodology for creating mathematical lumped models, decomposing a system into basic components represented by elementary physical laws and relationships expressed as equations.
Our approach is based on Modelica, an object-oriented, equation-based, visual, non-proprietary modeling language, together with Physiolibrary, an open-source library for the domain of physiology. Results: We demonstrate this methodology on an open implementation of a range of simple to complex cardiovascular models, with great complexity variance (simulation time from several seconds to hours).
The parts of different complexity could be combined together. Conclusions: Thanks to the equation-based nature of Modelica, a hierarchy of subsystems can be built with an appropriate connecting component.
Such a structural model follows the concept of the system rather than the computational order. Such a model representation retains structural knowledge, which is important for e.g., model maintainability and reusability of the components and multidisciplinary cooperation with domain experts not familiar with modeling methods.