Improving low-cycle fatigue life reliability sensitivity analysis of powder-metallurgy turbine disk under cyclic loading

Aim. The full paper starts with a review of relevant papers^[1-7]. In our opinion, the methods in these papers are either too tedious in computation or not sufficient in accuracy. We now propose a method based on the Markov-chain simulation that is, we believe, better. In the full paper, we explain our method in detail. In this abstract, we just list the two topics of explanation and add some pertinent remarks. The first topic is; the non-linear limit-state function for the reliability analysis of the low-cycle fatigue life of powder-metallurgy turbine disk under cyclic loading. In topic 1, we construct the limit-state function by using the interference model that takes into consideration both the required life and the actual life. The second topic is: the method for reliability sensitivity analysis of the low-cycle fatigue life of powder-metallurgy turbine disk. Its three subtopics are: the AFOSM (Advanced First Order and Second Moment) reliability sensitivity analysis (subtopic 2.1), the reliability sensitivity analysis based on the Monte-Carlo simulation (subtopic 2.2) and the reliability sensitivity analysis based on the Markov-chain simulation (subtopic 2.3). In topic 2, on the basis of non-linear limit-state function, three methods-approximate AFOSM analysis, Monte-Carlo simulation and Markov-chain simulation-are employed to analyze the reliability sensitivity of the low-cycle fatigue life of powder-metallurgy turbine disk. Finally we did numerical simulation, whose results are summarized in two tables and six figures in the full paper. These simulation results show preliminarily that: (1) the AFOSM method's computation is relatively not tedious but its accuracy is too low, (2) the Monte-Carlo simulation method is probably the most accurate but the computation required is far too tedious and (3) only our Markov-chain simulation method is not only relatively not tedious but also gives results that are very close to those obtained with the Monte-Carlo method.