Abstract
From the beginning, engineers have focused on the special case of determinism in the design process, and an enormous methodology has been developed to support this approach. Today, however, customers are demanding greater reliability and are imposing greater penalties for failure. In order to achieve higher reliability, and in order to asses risk of failure, probabilistic approaches will almost certainly have to be employed. While designers have always used probability in their work, it has usually been done with risk represented in a single factor of safety. This paper focuses on the application of probability theory to the design of high temperature components which are subjected to low cycle fatigue loads. Creep low cycle fatigue interaction and probabilistic design are both complex subjects. In order to make the probabilistic design of components subjected to creep and low cycle fatigue tractable, the calculation models must be as simple as possible without sacrificing too much on accuracy. In this paper, cumulative damage is determined using Miner’s Rule in conjunction with “range pair” cycle counting. The effect of operation at elevated temperatures is included using Coffin’s frequency modified approach. A first order second moment (FORM) method for including probabilistic effects is developed and some sample calculations are presented. It is shown that the traditional deterministic approach using a single factor of safety does not provide a uniform margin of safety for all design conditions.
Probabilistic design for reliability in electronics and photonics: Role, significance, attributes, challenges