[en] Mean stress effects significantly influence the fatigue life of components. In general, tensile mean stresses are known to reduce the fatigue life of components, whereas compressive mean stresses are known to increase it. To date, various methods that account for mean stress effects have been studied. In this research, considering the high accuracy of mean stress correction and the difficulty in obtaining the material parameter of the Walker method, a practical method is proposed to describe the material parameter of this method. The test data of various materials are then used to verify the proposed practical method. Furthermore, by applying the Walker material parameter and the Smith-Watson-Topper (SWT) parameter, a modified strain-life model is developed to consider sensitivity to mean stress of materials. In addition, three sets of experimental fatigue data from super alloy GH4133, aluminum alloy 7075-T651, and carbon steel are used to estimate the accuracy of the proposed model. A comparison is also made between the SWT parameter method and the proposed strainlife model. The proposed strain-life model provides more accurate life prediction results than the SWT parameter method.

[en] The value of exponent d in Corten-Dolan's model is generally considered to be a constant. Nonetheless, the results predicted on the basis of this statement deviate significantly from the real values. In consideration of the effects of damage and stress state on fatigue life prediction, Corten-Dolan's model is improved by redefining the exponent d used in the traditional model. The improved model performs better than the traditional one with respect to the demonstration of a fatigue failure mechanism. Predictions of fatigue life on the basis of investigations into three metallic specimens indicate that the errors caused by the improved model are significantly smaller than those induced by the traditional model. Meanwhile, predictions derived according to the improved model fall into a narrower dispersion zone than those made as per Miner's rule and the traditional model. This finding suggests that the proposed model improves the life prediction accuracy of the other two models. The predictions obtained using the improved Corten-Dolan's model differ slightly from those derived according to a model proposed in previous literature; a few life predictions obtained on the basis of the former are more accurate than those derived according to the latter. Therefore, the improved model proposed in this paper is proven to be rational and reliable given the proven validity of the existing model. Therefore, the improved model can be feasibly and credibly applied to damage accumulation and fatigue life prediction to some extent.

[en] Estimation of remaining useful life (RUL) is helpful to manage life cycles of machines and to reduce maintenance cost. Support vector machine (SVM) is a promising algorithm for estimation of RUL because it can easily process small training sets and multi-dimensional data. Many SVM based methods have been proposed to predict RUL of some key components. We did a literature review related to SVM based RUL estimation within a decade. The references reviewed are classified into two categories: improved SVM algorithms and their applications to RUL estimation. The latter category can be further divided into two types: one, to predict the condition state in the future and then build a relationship between state and RUL; two, to establish a direct relationship between current state and RUL. However, SVM is seldom used to track the degradation process and build an accurate relationship between the current health condition state and RUL. Based on the above review and summary, this paper points out that the ability to continually improve SVM, and obtain a novel idea for RUL prediction using SVM will be future works.

[en] A probabilistic Physics of Failure-based framework for fatigue life prediction of aircraft gas turbine discs operating under uncertainty is developed. The framework incorporates the overall uncertainties appearing in a structural integrity assessment. A comprehensive uncertainty quantification (UQ) procedure is presented to quantify multiple types of uncertainty using multiplicative and additive UQ methods. In addition, the factors that contribute the most to the resulting output uncertainty are investigated and identified for uncertainty reduction in decision-making. A high prediction accuracy of the proposed framework is validated through a comparison of model predictions to the experimental results of GH4133 superalloy and full-scale tests of aero engine high-pressure turbine discs. - Highlights: • A probabilistic PoF-based framework for fatigue life prediction is proposed. • A comprehensive procedure forquantifyingmultiple types of uncertaintyis presented. • The factors that contribute most to the resulting output uncertainty are identified. • The proposed frameworkdemonstrates high prediction accuracybyfull-scale tests.

[en] When a group of identical products is operating in field, the aggregation of failures is a catastrophe to engineers and customers who strive to develop reliable and safe products. In order to avoid a swarm of failures in a short time, it is essential to measure the degree of dispersion from different failure times in a group of products to the first failure time. This phenomenon is relevant to the crowding of system conditions near the worst one among a group of products. The group size in this paper represents a finite number of products, instead of infinite number or a single product. We evaluate the reliability of the product fleet from two aspects. First, we define near-extreme system condition and near-extreme failure time for offline solutions, which means no online observations. Second, we apply them to a continuous degradation system that breaks down when it reaches a soft failure threshold. By using particle filtering in the framework of prognostics and health management for a group of products, we aim to estimate near-extreme system condition and further predict the remaining useful life (RUL) using online solutions. Numerical examples are provided to demonstrate the effectiveness of the proposed method. - Highlights: • The aggregation of failures is measured for a group of identical products. • The crowding of failures is quantitated by the near-extreme evaluations. • Near-extreme system condition are given for offline solutions. • Near-extreme remaining useful time are provided for online solutions.