Model-Based Dependability Analysis of Physical Systems with Modelica. (8th March 2017)

• Record Type: Journal Article
• Authors: Tundis, Andrea; Buffoni, Lena; Fritzson, Peter; Garro, Alfredo
• Other Names: Ramírez Franco Academic Editor.
• Is Part Of: Modelling and simulation in engineering. Volume 2017(2017)
• Publication Date: 2017-03-08
• Subjects: Engineering -- Simulation methods -- Periodicals; Engineering -- Mathematical models -- Periodicals; 620.004
• DOI: 10.1155/2017/1578043
• Languages: English
• ISSNs: 1687-5591

Abstract

Abstract : Modelica is an innovative, equation-based, and acausal language that allows modeling complex physical systems, which are made of mechanical, electrical, and electrotechnical components, and evaluates their design through simulation techniques. Unfortunately, the increasing complexity and accuracy of such physical systems require new, more powerful, and flexible tools and techniques for evaluating important system properties and, in particular, the dependability ones such as reliability, safety, and maintainability. In this context, the paper describes some extensions of the Modelica language to support the modeling of system requirements and their relationships. Such extensions enable the requirement verification analysis through native constructs in the Modelica language. Furthermore, they allow exporting a Modelica-based system design as a Bayesian Network in order to analyze its dependability by employing a probabilistic approach. The proposal is exemplified through a case study concerning the dependability analysis of a Tank System.