Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Reliability investigation for load-sharing models applied to refrigeration systems
Submission Author:
Helio Fiori de Castro , SP , Brazil
Co-Authors:
tiago botega, Laís Bittencourt Visnadi, Helio Fiori de Castro
Presenter: tiago botega
doi://10.26678/ABCM.COBEM2023.COB2023-1329
Abstract
A method to increase engineering systems' reliability is using redundant components. In many of these systems, the system load is shared by their components when operating simultaneously. Therefore, when a component fails, the system continues to function, but the load on the remaining components becomes greater, which leads to an increase in the failure rate of these components. To model the reliability of these systems and, thus, estimate their failure probability, a model based on a stochastic process may be used. Stochastic processes can be defined as a sequence of random variables indexed to time and events. A particular case of a stochastic process is the Markov Chain, whose main property is the absence of memory. This property establishes that the behavior of future probabilities is determined only by the current state, widely applied to engineering and reliability problems. Aiming to study models of load-sharing and reliability in redundant mechanical components, a numerical model based on Markov Chains was implemented to describe the reliability of components and systems using two different load-sharing models to calculate failure rates. An industrial application in refrigeration systems with compressors working in redundancy was analyzed. From the results obtained via simulation, it is noted that the studied models are efficient in describing the load-sharing behavior in refrigeration systems with more than one compressor operating in redundancy, estimating load-sharing parameters, and calculating system reliability.
Keywords
reliability, Load-sharing, Markov chain
