Modelling the dependency of the Smith–Watson–Topper parameter on the cycles‐to‐failure using serial hybrid neural networks

Calculating the fatigue damage with a strain‐based approach requires an ?–N durability curve that links the strain amplitude to the corresponding number of cycles‐to‐failure. This ?–N curve is usually modelled by the Coffin–Manson relationship. If a loading mean‐level also needs to be considered, the original Coffin–Manson relationship is modified using a Smith–Watson–Topper parameter. In this article a methodology for modelling the dependence of the Smith–Watson–Topper parameter on the number of cycles‐to‐failure is presented. The core of the presented methodology represents a multilayer perceptron neural network combined with the Smith–Watson–Topper analytical model. The article presents the theoretical background of the methodology, which is applied for the case of the experimental fatigue data. The results show that it is possible to model ?–N curves for different influential parameters, such as the specimen's diameter and the testing temperature. The results further show that it is possible to predict ?–N curves even for those combinations of the influential parameters for which no experimental data about the material endurance is available. This fact makes the presented model very suitable for the application in an R&D process when a durability of a product should be estimated on the basis of a very limited set of experimental data about the material endurance characteristics.     

Application of Bayesian Methods for Age‐Dependent Reliability Analysis

In this article, the authors present a general methodology for age‐dependent reliability analysis of degrading or ageing components, structures and systems. The methodology is based on Bayesian methods and inference—its ability to incorporate prior information and on ideas that ageing can be thought of as age‐dependent change of beliefs about reliability parameters (mainly failure rate), when change of belief occurs not only because new failure data or other information becomes available with time but also because it continuously changes due to the flow of time and the evolution of beliefs. The main objective of this article is to present a clear way of how practitioners can apply Bayesian methods to deal with risk and reliability analysis considering ageing phenomena. The methodology describes step‐by‐step failure rate analysis of ageing components: from the Bayesian model building to its verification and generalization with Bayesian model averaging, which as the authors suggest in this article, could serve as an alternative for various goodness‐of‐fit assessment tools and as a universal tool to cope with various sources of uncertainty. The proposed methodology is able to deal with sparse and rare failure events, as is the case in electrical components, piping systems and various other systems with high reliability. In a case study of electrical instrumentation and control components, the proposed methodology was applied to analyse age‐dependent failure rates together with the treatment of uncertainty due to age‐dependent model selection. Copyright © 2013 John Wiley & Sons, Ltd.     

Stepped‐isothermal fatigue analysis of engine piston

Stepped‐isothermal fatigue failure is the main failure mechanism of modern engine pistons under bench reliability test condition. This paper presents a methodology for stepped‐isothermal fatigue analysis of engine pistons, which consists of a fatigue criterion, evaluation of temperature and stress distribution by finite element analysis and the final life prediction. The major character of the methodology is the fatigue definition of engine pistons with respect to engine load change cycle and a damage‐based fatigue criterion accounting for the nonlinear creep–fatigue damage. Taking as an example, the fatigue life of an engine piston was predicted by the proposed analysis procedures. The analysis results showed that the most critical area was located in the throat edge. Moreover, the proposed methodology can give a relatively accurate and reasonable life prediction for an engine piston under the loading condition of bench reliability test, with a benefit of decreasing the needed component's reliability tests and design time.     

Sustainable Manufacturing Oriented Prognosis for Facility Reuse Combining ANN and Reliability

Reuse is considered as one of the most reasonable strategies in realizing sustainability, because it enables longer useful life of facilities. This article presents an effective methodology of artificial neural network–based prognosis combined with reliability methods to evaluate and guarantee the reusability of a facility. The methodology provides the assessment of the degradation trend and prediction of the remaining life of facilities based on online condition monitoring data and historical data utilizing back propagation artificial neural networks. In addition, the corresponding reliability of a facility is calculated by fitting suitable life distribution against the in‐house time‐to‐failure data. Furthermore, maintenance decision is made by predicting the time when reliability or remaining life of a facility reaches the threshold, as determined by the facility's reusability. Application results show that the proposed methodology provides sufficient condition information for reuse decision making from both historical and online perspectives; a facility can be reused for many times during its lifetime until its reuse is no longer economic, which can assist in the achievement of the goal of manufacturing with fewer resources and assets. Copyright © 2011 John Wiley & Sons, Ltd.     

Failure prediction and strength improvement of uni-directional composite single lap bonded joints

A methodology is presented for the failure prediction of the composite single lap bonded joints considering both the composite adherend and the bondline failures. An elastic-perfectly plastic model of the adhesive and a delamination failure criterion were used in the methodology. The failure predictions using the finite element analysis and the proposed methodology were performed. The failure prediction results such as failure mode and strength showed very good agreements with the test results for the joint specimens with various bonding methods and parameters. Based on the numerical investigation, the optimal joint strength condition was found and a new joint strength improvement technique was suggested. The suggested technique was verified to have a significant effect on the joint strength improvement.     

Integrated System Health Management‐based Fuzzy On‐board Condition Prediction for Manned Spacecraft Avionics

The purpose of this study is to develop a prediction methodology for a condition assessment of on‐board integrated health management systems in manned spacecraft avionics. The framework of this study is based on two main premises. The first is the need to examine the problems in the on‐board prediction tools in space avionics integrated system health management, an area that has been rarely focused on. The second is the need to consider the failure correlation coefficients to enable accurate health predictions. To deal with the uncertainty in the prediction process, a fuzzy theory–gray model–support vector machine approach, which uses fuzzy theory combined with a gray model and a support vector machine, is used to make the prediction. An example is given to demonstrate the accuracy and reliability of the model. Copyright © 2014 John Wiley & Sons, Ltd.     

Two‐level multiscale enrichment methodology for modeling of heterogeneous plates

A new two‐level multiscale enrichment methodology for analysis of heterogeneous plates is presented. The enrichments are applied in the displacement and strain levels: the displacement field of a Reissner–Mindlin plate is enriched using the multiscale enrichment functions based on the partition of unity principle; the strain field is enriched using the mathematical homogenization theory. The proposed methodology is implemented for linear and failure analysis of brittle heterogeneous plates. The eigendeformation‐based model reduction approach is employed to efficiently evaluate the non‐linear processes in case of failure. The capabilities of the proposed methodology are verified against direct three‐dimensional finite element models with full resolution of the microstructure. Copyright © 2009 John Wiley & Sons, Ltd.     

A hierarchical combinatorial reliability model for smart home systems

As an application of the Internet of Things, smart home systems have received significant attentions in recent years due to their precedent advantages, eg, in ensuring efficient electricity transmission and integration with renewable energy. This paper proposes a hierarchical and combinatorial methodology for modeling and evaluating reliability of a smart home system. Particularly, the proposed methodology encompasses a multi‐valued decision diagram‐based method for addressing phased‐mission, standby sparing, and functional dependence behaviors in the physical layer; and a combinatorial procedure based on the total probability theorem for addressing probabilistic competing failure behavior with random propagation time in the communication layer. The methods are applicable to arbitrary types of time‐to‐failure and time‐to‐propagation distributions for system components. A detailed case study of an example smart home system is performed to demonstrate applications of the proposed method and effects of different component parameters on the system reliability.     

Accelerating cyclic plasticity simulations using an adaptive wavelet transformation based multitime scaling method

Cyclic finite element simulations of complex materials, for example, polycrystalline metals, are widely used to study fatigue failure due to plasticity and damage. Typically, this requires the simulation of a large number of cycles to failure for accurate determination of evolving deformation variables. Modeling cyclic deformation using conventional methods of time integration in semidiscretization techniques can however be computationally challenging. Single time scale integration methods typically follow the high frequency characteristics and discretize each cycle into a number of time steps over which integration is performed. To overcome this computational challenge, the wavelet transformation‐based multitime scale (WATMUS) method proposed in an earlier work by the authors is advanced and validated in this paper to perform accelerated finite element simulations of materials undergoing rate‐dependent plasticity for large number of cycles. Specifically, the WATMUS algorithm is integrated with crystal plasticity finite element method to perform accelerated simulations of polycrystalline alloys. The WATMUS methodology is also endowed with adaptive capabilities to optimally construct the wavelet basis functions and determine coarse‐scale cycle steps. Accuracy and efficiency of the WATMUS methodology is conclusively demonstrated by comparing the results with cyclic single‐time scale crystal plasticity finite element simulations performed on image‐based microstructure of titanium alloys. Copyright © 2012 John Wiley & Sons, Ltd.     

Prediction intervals for load-sharing systems in accelerated life testing

Based on accelerated lifetime experiments, we consider the problem of constructing prediction intervals for the time point at which a given number of components of a load-sharing system fails. Our research is motivated by lab experiments with prestressed concrete beams where the tension wires fail successively. Due to an audible noise when breaking, the time points of failure could be determined exactly by acoustic measurements. Under the assumption of equal load sharing between the tension wires, we present a model for the failure times based on a birth process. We provide a model check based on a Q-Q plot including a simulated simultaneous confidence band and four simulation-free prediction methods. Three of the prediction methods are given by confidence sets where two of them are based on classical tests and the third is based on a new outlier-robust test using sign depth. The fourth method uses the implicit function theorem and the δ-method to get prediction intervals without confidence sets for the unknown parameter. We compare these methods by a leave-one-out analysis of the data on prestressed concrete beams. Moreover, a simulation study is performed to discuss advantages and drawbacks of the individual methods.     

Reliability‐Based Design Optimization Considering Probabilistic Degradation Behavior

This article proposes a reliability‐based design optimization methodology by incorporating probabilistic degradation in the fatigue resistance of material. The probabilistic damage accumulation is treated as a measure of degradation in the fatigue resistance of material and modeled as nonstationary probabilistic process to capture the time‐dependent distribution parameters of damage accumulation. The proposed probabilistic damage accumulation model is then incorporated into reliability‐based design optimization model by building a dynamic reliability model inferred from the stress–strength interference model. The proposed approach facilitates to capture the dynamic degradation behavior while optimizing design variables at an early design stage to improve the overall reliability of product. The applicability of the proposed approach is demonstrated using suitable examples. Copyright © 2012 John Wiley & Sons, Ltd.     

A multi-temporal scale approach to high cycle fatigue simulation

High cycle fatigue (HCF) is a failure mechanism that dominates the life of many engineering components and structures. Time scale associated with HCF loading is a main challenge for developing a simulation based life prediction framework using conventional FEM approach. Motivated by these challenges, the extended space–time method (XTFEM) based on the time discontinuous Galerkin formulation is proposed. For HCF life prediction, XTFEM is coupled with a two-scale continuum damage mechanics model for evaluating the fatigue damage accumulation. Direct numerical simulations of HCF are performed using the proposed methodology on a notched specimen of AISI 304L steel. It is shown the total fatigue life can be accurately predicted using the proposed simulation approach based on XTFEM. The presented computational framework can be extended for predicting the service and the residual life of structural components.     

A two‐dimensional BEM/FEM coupling applied to viscoelastic analysis of composite domains

A coupling between the boundary‐element and finite‐element methods is studied for the viscoelastic analysis of reinforced media. The viscous behaviour of the composed body is taken into account by an alternative BEM methodology developed for the Boltzmann model. This methodology is based on differential constitutive relations for viscoelasticity. The reinforcements are modelled by finite elements and are considered elastic. The coupling is based on the sub‐region technique due to its generality and easy implementation. The resulting time‐marching process is able to represent both the instantaneous and the time‐dependent behaviour of a body subjected to general boundary conditions. The method is validated by an experimental result and its accuracy tested by comparing numerical results with analytical solutions. The generality of the method is proved by an infinite domain application. Copyright © 2003 John Wiley & Sons, Ltd.     

An efficient fatigue and creep‐fatigue life prediction method by using the hysteresis energy density rate concept

Fatigue damage, time‐dependent creep damage and their interaction are considered as the main failure mechanisms for many high temperature structural components. A generalized methodology for predicting both the high temperature low cycle fatigue (HTLCF) and creep‐fatigue lives by using the hysteresis energy density rate (HEDR) and fatigue damage stress concepts was proposed. Experimental data for HTLCF and creep‐fatigue in Alloy 617, Haynes 230 and P92 steel were respectively collected to validate the method. A better prediction capacity and most of the data points that fall within a 1.5 scatter band were obtained compared with the traditional energy‐based method, time fraction rule and ductility exhaustion model. Moreover, a creep‐fatigue damage diagram was also constructed by using the proposed approach.     

Ensemble Support Vector Machine Algorithm for Reliability Estimation of a Mining Machine

In this study, a support vector machine (SVM)‐based ensemble model was developed for reliability forecasting. The hyperparameters of the SVM were selected by applying a genetic algorithm. Input variables of the SVM model were selected by maximizing the mean entropy value. The diverse members of the ensemble model were obtained by a k‐means clustering algorithm, and one ensemble member was selected from each cluster by choosing the closest from the cluster center. The optimum cluster number was selected using the Davies–Bouldin index. The developed model was validated by a benchmark turbocharger data set. A comparative study reveals that the proposed method performs better than existing methods on benchmark data sets. A case study was conducted investigating a dumper operated at a coal mine in India. Time‐to‐failure historical data for the dumper were collected, and cumulative time to failure was calculated for reliability forecasting. Study results demonstrate that the developed model performs well with high accuracy (R² = 0.97) in the prediction of dumper failure, and a comparison with other methods demonstrates the superiority of the proposed ensemble SVM model. Copyright © 2014 John Wiley & Sons, Ltd.     

Simulation of multi‐directional crack growth in braided composite T‐piece specimens using cohesive models

This paper presents an approach to the prediction of multi‐directional crack growth in braided composite T‐piece specimens using cohesive zone models. Failure predictions given by the cohesive model together with experimental results are presented. The effects of single and multi‐directional crack model on failure prediction are investigated. Results show that both models agree with experimental data, single‐crack model agrees with the maximum value of experimental data, and multi‐directional crack model agrees with the mean value. The simulation of multi‐directional crack growth of T‐piece specimens is important in the study of their failure mechanism. Investigation reported in this paper indicates that prediction of failure loads by the multi‐directional crack model is necessary in the design of some particular composite components which have similar features to the T‐piece specimens.     

A practical approach for predicting fatigue reliability under random cyclic loading

The purpose of this paper is to provide a simple approach for reliability analysis based on fatigue or overstress failure modes of mechanical components, and explain how this integrated method carries out spectral fatigue damage and failure reliability analysis. In exploring the ability to predict spectral fatigue life and assess the reliability under a specified dynamics environment, a methodology for reliability assessment and its corresponding fatigue life prediction of mechanical components using a supply-demand interference approach is developed in this paper. Since the methodology couples dynamics analysis and stochastic analysis for fatigue damage and reliability prediction, the conversion of the duty cycle history for the reliability study of an individual component is also presented. Using the proposed methodology, mechanical component reliability can be predicted according to different mission requirements. For an explanation of this methodology, a probabilistic method of deciding the relationship between the allowable stress or fatigue endurance limit and reliability is also presented.     

Optimal transportation meshfree method in geotechnical engineering problems under large deformation regime

Meshfree methods have been demonstrated as suitable and strong alternatives to the more standard numerical schemes such as finite elements or finite differences. Moreover, when formulated in a Lagrangian approach, they are appropriate for capturing soil behavior under high‐strain levels. In this paper, the optimal transportation meshfree method has been applied for the first time to geotechnical problems undergoing large deformations. All the features employed in the current methodology (ie, F‐bar, explicit viscoplastic integration, and master‐slave contact) are described and validated separately. Finally, the model is applied to the particular case of shallow foundations by using von Mises and Drucker‐Prager yield criteria to find the load at failure in the. The presented methodology is demonstrated to be robust and accurate when solving this type of problems.     

A micromechanics-based strength prediction methodology for notched metal-matrix composites

An analytical micromechanics-based strength prediction methodology was developed to predict failure of notched metal-matrix composites. The stress/strain behaviour and notched strength of two metal-matrix composites, boron/aluminium (B/Al) and silicon carbide/titanium (SCS-6/Ti-15-3), were predicted. The prediction methodology combines analytical techniques ranging from a three-dimensional finite element analysis of a notched specimen to a micromechanical model of a single fibre. In the B/Al laminates, a fibre failure criterion based on the axial and shear stress in the fibre accurately predicted laminate failure for a variety of lay-ups and notch-length-to-specimen-width ratios with both circular holes and sharp notches when matrix plasticity was included in the analysis. For the SCS-6/Ti-15-3 laminates, a fibre failure criterion based on the axial stress in the fibre correlated well with experimental results for static and post-fatigue residual strengths when fibre/matrix debonding and matrix cracking were included in the analysis. The micromechanics-based strength prediction methodology presented here offers a direct approach to strength prediction by modelling behaviour and damage on a constituent level, thus explicitly including matrix non-linearity, fibre/matrix interface debonding and matrix cracking.