Simulation of bearing degradation by the use of the gamma stochastic process

Abstract

An effective predictive maintenance reposed on modeling, simulation, and on supervisory and prognostic techniques used to model the various phenomena. On this basis, and based on significant knowledge and parameters, we propose an approach based on stochastic processes that represent a mathematical structure for simulation, mainly the processes of continuous degradation and more particularly the Gamma process. Our work is devoted to the monitoring of the degradation process of the bearings at the level of a motor pump and makes it possible to evaluate the limiting operating time, as well as the evolution in time of the change of state. This methodology allows us to develop a mathematical model that describes the process of bearing degradation, thus providing a good prediction of failures and efficient maintenance planning for systems whose behavior is only partially predictable.