RELIABILITY MODEL FOR A NONLINEAR DAMAGE PROCESS.

Several types of damage processes lead to the failure of mechanical devices. Some examples of these damage processes are: wear, fatigue, creep and corrosion. We consider a generalized damage process defined by the random function D(t); D(t) equals At**B, t is an element of T, where A, B and T are random variables, and D(t) equals damage incurred at time t. If the life of the device is terminated when the damage exceeds a limiting value D//l (i. e. D(t) greater than equivalent to D//l), then it is shown that, under certain assumptions regarding random variables A and B, the reliability model of the device is expressed by a given equation. Various statistical characteristics of the model are discussed. The generalized nature of the proposed model is emphasized, and it is shown that the lognormal model is a special case of the proposed model. A method of estimation of the parameters of the reliability model from a set of realizations (sample functions) of the damage process is illustrated.