PREDICTION DU CHANGEMENT D'ETAT DE LA VISCOSITE DU LUBRIFIANT PAR LES CHAINES DE MARKOV

Abstract

The quality of lubricating oil used in the engine plays an important role in extending engine life and improving engine performance. Viscosity is one of the major physicochemical factors of lubricant quality and effectiveness, Its ability to provide the effective thickness of the lubricating layer between the friction surfaces, which prevents severe wear and failure of engines. In this study, we presented the methodology which makes it possible to model the degradation of the viscosity of lubricating oils used in combustion engines, and which are operated in very severe conditions. For each oil studied, we propose to associate a Markov chain describing the transition of its viscosity from one state (i) to another state, Knowing that the viscosity of engine oils passes through the following states: good, admissible, degraded, very degraded. We have confirmed that the proposed model gives detailed results describing the different states of the oil analyzed.