From Modelica models to dependability analysis

Abstract

Modelica is a modeling language which was created in order to ease the description of multi physics systems thanks to an object oriented approach. Modelica models usually represent only the nominal functioning of systems and are used to simulate them for design purposes. This article proposes various ways to derive dependability models from such Modelica models, as automatically as possible. Depending on the tightness of coupling between the continuous processes and discrete events such as failures and repairs, the methods proposed here range from the addition of stochastic behavior in the Modelica model itself to the association of the system structure to a library written in the Figaro reliability modeling language. Such an association allows the use of the Figaro workbench tools, specially designed for dependability analysis; it is then possible to generate a fault tree, or to create automatically a discrete stochastic simulation model. All these possibilities are exemplified through two classical use cases: a telecommunication network, and a "heated tank" system.

A Model-Based Hybrid Approach for Circuit Breaker Prognostics Encompassing Dynamic Reliability and Uncertainty

“…Instead of considering a priori established failure specifications of components, these approaches add the possibility to specify components with a variable failure rate which depends on the operating conditions of the system. This community has also started to create design tools to create hybrid models from user-friendly specifications, e.g., implementing PDMPs in Python [32] or linking reliability analysis and multi-physics specification tools [33]. There are other techniques which can also be used to solve dynamic reliability problems such as Dynamic Bayesian Networks [34], Stochastic Activity Networks [35], or Fluid Stochastic Petri Nets [36].…”

“…Dynamic reliability approaches focus on updating dynamically the probability density function representing the system failure state according to operational conditions (e.g., [30]- [33]). For instance, the Weibull distribution allows the specification of a time-varying failure rate of the system (λ(t)) with the following density function:…”

Model-Based Dependability Analysis of Physical Systems with Modelica

“…The engineering discipline which aims at providing an integrated and methodological approach to deal with system dependability is commonly indicated by the acronym RAMS (Reliability, Availability, Maintainability, and Safety), whereas the main objective to identify causes and consequences of system failures is called RAMS analyses [38][39][40]. Facing dependability challenges in Modelica world is a quite young research topic as proved by some recent research works [41]. Indeed, some research efforts are already available in literature for enabling the dependability analysis of system based on the Modelica language and related tools.…”

Future Challenges in Mechatronics

“…In the frame of the MODRIO project [31], a prototype was developed to automatically generate Figaro models from Modelica [32] models and a Figaro knowledge base [33]. Modelica specific constructs are used to declare the correspondence with a Figaro block from the knowledge base and other necessary information.…”