Predictive control of electrical equipment reliability on the basis of the non-linear canonical model of a vector random sequence

Abstract

Method of individual prediction of electrical equipment reliability on the basis of the analysis of the probability of no-failure operation at future moments of time is offered. A posteriori probability of no-failure operation is calculated using the statistical modeling of the realizations of a vector random sequence of the change of the parameters of electrical device in the prediction area. Relative frequency of the number of realizations belonging to the area of allowable values of the parameters of an electrical device is taken as the estimation of a sought a posteriori probability. Statistical modeling is performed on the basis of non-linear canonical expansion of a random sequence. Taking into full account of stochastic relations and properties of electric equipment parameters is a substantial advantage of this mathematical model which will allow to achieve maximum accuracy of the solving of the problem of predictive control of reliability. © 2019 IEEE.