Reliability Analysis Framework Research Articles

A mission profile-based reliability analysis framework for photovoltaic DC-DC converters

Reliability of DC-DC converters is important in photovoltaic (PV) applications like building integrated PV systems, where the module-level converter may be stressed significantly. Understanding and predicting the most life-limiting components with accurate degradation models in such systems enable the design for reliability. In this paper, a mission profile-based reliability analysis framework for PV DC-DC converters is proposed where the inputs and models of the framework can be adjusted according to the converter topology, the components and the failure mechanisms under investigation. The framework is demonstrated by comparing the influence of two yearly mission profiles on the solder joint degradation of a MOSFET in an interleaved boost converter. This is done by using an electro-thermal circuit simulation in PLECS and a finite element MOSFET model in COMSOL. This framework allows for exploring more accurate models or even simplifying parts with low sensitivity in order to obtain a thorough understanding of their accuracy and to determine the overall converter reliability.

Reliability-Based Serviceability Limit State Design of a Jacket Substructure for an Offshore Wind Turbine

The development of a structurally optimized foundation design has become one of the main research objectives for offshore wind turbines (OWTs). The design process should be carried out in a probabilistic way due to the uncertainties involved, such as using parametric uncertainties regarding material and geometric properties, and model uncertainties in resistance prediction models and regarding environmental loads. Traditional simple deterministic checking procedures do not guarantee an optimized design because the associated uncertainties are not fully considered. In this paper, a reliability analysis framework is proposed to support the optimized design of jacket foundations for OWTs. The reliability analysis mainly considers the serviceability limit state of the structure according to the requirements of the code. The framework consists of two parts: (i) an important parameter identification procedure based on statistical correlation analysis and (ii) a finite element-simulation-based reliability estimation procedure. The procedure is demonstrated through a jacket structure design of a 3 MW OWT. The analysis results show that the statistical correlation analysis can help to identify the parameters necessary for the overall structural performance. The Latin hypercube sampling and the Monte Carlo simulation using FE models effectively and efficiently evaluate the reliability of the structure while not relying on a surrogate limit state function. A comparison between the proposed framework and the deterministic design shows that the framework can help to achieve a better result closer to the target reliability level.

An efficient surrogate-aided importance sampling framework for reliability analysis

Surrogates in lieu of expensive-to-evaluate performance functions can accelerate the reliability analysis greatly. This paper proposes a new two-stage framework for surrogate-aided reliability analysis named Surrogates for Importance Sampling (S4IS). In the first stage, a coarse surrogate is built to gain the information about failure regions. The second stage zooms into the important regions and improves the accuracy of the failure probability estimator by adaptively selecting support points. The learning functions are proposed to guide the selection of support points such that the exploration and exploitation can be dynamically balanced. As a generic framework, S4IS has the potential to incorporate different types of surrogates (Gaussian Processes, Support Vector Machines, Neural Network, etc.). The effectiveness and efficiency of S4IS are validated by five illustrative examples, which involve system reliability, highly nonlinear limit-state functions, small failure probability and moderately high dimensionality. The implementation of S4IS is made available to download at https://sites.google.com/site/josephsaihungcheung/.

Epistemic uncertainty-based reliability analysis for engineering system with hybrid evidence and fuzzy variables

With the rapid development of reliability analysis theory, the safety assessment for engineering systems with hybrid epistemic uncertainties has received increasing attentions. To overcome the imperfection of traditional single-uncertainty modeling methods, this paper introduces evidence variables and fuzzy variables simultaneously to describe the epistemic uncertain parameters. A novel dual-stage reliability analysis framework is presented, where the first stage incorporates the evidence information by the belief and plausibility measures; the second stage incorporates the fuzzy information by a membership function-like formula. To improve the computational efficiency of response bound prediction associated with various joint focal elements and λk-cut interval variables, a universal metamodel is constructed by radial basis functions in the global design space. The Latin Hypercube design method without overlapped data is adopted as the sampling strategy. Finally, two numerical examples verify the effectiveness of the proposed method in both mathematical theory and engineering application.

Two-stage dimension reduction method for meta-model based slope reliability analysis in spatially variable soils

The traditional slope reliability methods may not efficiently carry out slope reliability considering inherent spatial variability (ISV) of soil properties. This paper aims to develop a two-stage dimension reduction method for efficient slope reliability analysis considering ISV of soil properties. First, the framework of slope reliability analysis based on surrogate model is provided. Second, the two-stage dimension reduction method is proposed systematically. Thereafter, the implementation procedure for slope reliability analysis using the proposed method is summarized. The validity of two-stage dimension method is illustrated with a multiple-layered soil slope. The results indicate that the proposed two-stage dimension method can reduce the number of original model evaluations significantly and improve the efficiency of slope reliability analysis considerably. The surrogate model based on the proposed method is proved to be unbiased. The loss of accuracy in modeling ISV has a slight influence on the minimum factor of safety, sliding path and sliding volumes of slope. The proposed method can yield sufficiently accurate reliability results, which is more efficient than direct Monte Carlo simulation. For the multi-layered soil slope reliability considering ISV of soil properties examined in this study, the proposed method performs well in terms of accuracy and efficiency.

Multiscale surrogate-based framework for reliability analysis of unidirectional FRP composites

In this paper, a Finite Element-based surrogate model is developed to efficiently estimate stiffness properties of unidirectional composite laminas, while accounting for geometric and material property uncertainties at micro, meso and laminate scale, all within a probabilistic framework. In the multi-scale build-up nature of composites, uncertainties occur in material properties and geometric characteristics. These uncertainties present a challenge in estimating composite material properties. The currently available property estimation/homogenisation tools are mainly divided into two categories: analytical methods constrained by configuration assumptions, and numerical homogenisation using Finite Element Analysis (FEA). The latter is more flexible and accurate, but computationally expensive. Hence, this paper develops a surrogate model based on a limited number of experimental FEA data points. Additionally, a transition phase is developed between micro and laminate scales that enables modelling of spatially varying uncertainties. As a result, this framework significantly decreases analysis duration compared with FEA techniques, and, because it is derived from FEA data points, can accurately represent a wide range of uncertainties.

Predictable Behavior of GFRP-Reinforced Bridge Decks: Formulation of a Strain-Based Capacity Model

This paper proposes a reliability analysis framework for glass fiber-reinforced polymer- (GFRP-) reinforced concrete systems with uncertain capacities and demands over time. Unfortunately, there has been limited discussion or research done related to the potential change of failure modes over time. Therefore, a rational approach is needed to integrate multiple failure modes in a single analysis framework, considering uncertainties of time-variant demands and capacities. To account for multiple failure modes, this study proposes the limit state function to estimate the safety margin, based on strain values of GFRP-reinforcing bars. A proposed limit state function can capture the likelihood of both shear and flexural failure modes, simultaneously. In this study, seven typical bridge deck configurations (e.g., varied deck thickness, girder spacing, and bar size) were exposed to various ambient temperatures. Simulation results show that reliability indices of 100-year exposure exhibit significant variance, ranging from 2.35 to 0.93, with exposure temperatures ranging from 13 to 33°C. Exposure temperature and time are the dominant factors influencing the reliability indices, so are the ones that need to be changed. As exposure time and/or exposure temperature increase, the flexural capacity model plays an important role to determine the reliability indices. When flexural and shear failure modes are equally dominant, reliability indices can capture risks of both failures, using the proposed strain-based approach.

Quantification of a human reliability analysis method for radiotherapy applications based on expert judgment aggregation

This paper develops the quantification framework of a new Human Reliability Analysis (HRA) method, for application to the radiotherapy domain. The method's reference data is obtained via expert judgment, due to the lack of domain-specific data. To avoid shortcomings of directly eliciting probabilities, experts are asked to assess the importance of specific factors for the failure probability, elicited on a qualitative scale. Each assessment is converted into statements about the order of magnitude of the probability value. The values are combined via an expert aggregation method, developed specifically for HRA. The paper includes an attempt to validate the elicitation against applicable Human Error Probability (HEP) values from existing HRA methods (thus, “convergent validation”). Indeed, some tasks are generic in nature and data can be assumed to be sector-independent (e.g. checking activities, interacting with interfaces, simple tasks such as identifying objects or characters/numbers). Differences in the values are identified and, when possible, linked to differences in the performance context characteristic of the field of application of the different methods.

A framework for fatigue reliability analysis of high-pressure turbine blades

Fatigue evolution under continued stresses is a process of degradation of material performance with many uncertainties. In order to quantify the uncertainties of materials and working conditions, a probabilistic method is utilized to estimate the reliability of structures by considering scatter of the fatigue life prediction model, in which improvements are provided to model the accumulation of the damage. Firstly, the fatigue parameters are modeled by the Bayesian theory and the finite element analysis. Secondly, the distributions of parameters are transformed by the probabilistic method into the distribution of fatigue life by using the fatigue life prediction model, and a damage accumulation model is chosen to characterize regulation evolution of properties. Finally, the probability distribution function transformation approach is employed to expound distribution of fatigue damage by the known distribution of fatigue life, and a general probabilistic method is then used to estimate the reliability. By combining the above methods, the framework for reliability analysis is established and then is used to calculate the reliability for high-pressure turbine blades in a low cycle fatigue region under variable amplitude loadings.

Improved reliability of stormwater detention basin performance through water quality data-informed real-time control

Stormwater detentions basins are designed to capture stormwater to reduce and delay peak flows and to improve water quality. A novel technology proposed to improve basin performance is real-time, active control of the basin outflow, in so-called “smart” stormwater systems. Existing studies demonstrate the performance of active controls that respond in real-time to basin water level, detention time, and rainfall forecast for one or a small number of rainfall events. We hypothesize that the performance of these active controls can be improved by incorporating real-time water quality data into the control algorithm. In addition, we hypothesize that active control performance depends on hydrologic variability, specifically the frequency and intensity of runoff inputs. In this paper, we test these hypotheses using a numerical modeling framework for systems-level reliability analysis of active and passive stormwater basin outflow control using a Monte Carlo method. The analysis is performed using the urban hydrology model EPA-SWMM driven by stochastic rainfall time-series generated from the Modified Bartlett-Lewis Rectangular Pulses Model. Water quality-informed real-time active control algorithms are developed, tested, and shown to display an improvement over traditional, passive (no control) systems and other storage-based active controls for water and pollutant capture. Seasonal and duration curve analysis showed that water level- and water quality- informed control performance varied for different storm return periods and this variability could partly be attributed to the fraction of time the valve is closed. In addition, control performance was sensitive to rainfall variability, generally decreasing as storms become less frequent and more intense. Therefore, control system performance may depend on seasonal and longer time-scale variability in climate and rainfall-runoff processes. We anticipate this study to be a starting point to incorporate theories of reliability to assess detention basin and conveyance network performance under more complex real-time control algorithms and failure modes.

A Comprehensive Reliability Analysis Framework for NTC Caches: A System to Device Approach

Near threshold computing (NTC) has significant role in reducing the energy consumption of modern very large scale integrated circuits designs. However, NTC designs suffer from functional failures and performance loss. Understanding the characteristics of the functional failures and variability effects is of decisive importance in order to mitigate them, and get the utmost NTC benefits. This paper presents a comprehensive cross-layer reliability analysis framework to assess the effect of soft error, aging, and process variation in the operation of near threshold voltage caches. The objective is to quantify the reliability of different SRAM designs, evaluate voltage scaling potential of caches, and to find a reliability-performance optimal cache organization for an NTC microprocessor. In this paper, the soft error rate (SER) and static noise margin (SNM) of 6T and 8T SRAM cells and their dependencies on aging and process variation are investigated by considering device, circuit, and architecture level analysis. Their experimental results show that in NTC, process variation and aging-induced SNM degradation is $2.5{\boldsymbol \times }$ higher than in the super threshold domain while SER is $8{\boldsymbol \times }$ higher. At NTC, the use of 8T instead of 6T SRAM cells can reduce the system-level SNM and SER by 14% and 22%, respectively. Besides, we observe that we can find the right balance between performance and reliability by using an appropriate cache organization at NTC which is different from the super threshold.

The Robust Architecture Based Reliability Analysis Framework of Complex Software System Using In-Degree and Out-Degree

The Robust Architecture Based Reliability Analysis Framework of Complex Software System Using In-Degree and Out-Degree

Time‐Dependent Reliability Analysis of RC Deep Beams considering Linear/Nonlinear Creep and Shrinkage Using ANFIS Network and MCS

This paper comprehensively studies the effects of linear and nonlinear creep and shrinkage on shear strength and deformation of reinforced concrete (RC) deep beams through a time‐dependent reliability analysis framework. The three‐dimensional finite element‐based program of ABAQUS was used to model RC deep beams. A viscoelastic approach was adopted to model linear creep and shrinkage by a user‐defined material model developed and implemented in user subroutines UMAT and UEXPAN. The software was initially examined using two experimental results for short/long‐term behavior of shallow and deep concrete beams. In the nonlinear range, creep was taken into account by the affinity hypothesis theorem to consider the effect of high‐level sustained loading. Due to the complicated and time‐consuming nature of the finite element method (FEM), adaptive neuro‐fuzzy inference system (ANFIS) and Monte Carlo simulation (MCS) replaced these complex analyses. Finally, based on the numerical results obtained from the analyses of case study samples, it was concluded that, in a serviceability limit state, for RC deep beams, the probability of failure was reduced to less than one‐fifth that for the shallow beams. On the contrary, in a strength limit state, a safety factor of about 1.7∼1.8 could be considered for the effect of sustained high‐level loading on the shear strength of RC deep beams.

SyRA: Early System Reliability Analysis for Cross-Layer Soft Errors Resilience in Memory Arrays of Microprocessor Systems

Cross-layer reliability is becoming the preferred solution when reliability is a concern in the design of a microprocessor-based system. Nevertheless, deciding how to distribute the error management across the different layers of the system is a very complex task that requires the support of dedicated frameworks for cross-layer reliability analysis. This paper proposes SyRA, a system-level cross-layer early reliability analysis framework for radiation induced soft errors in memory arrays of microprocessor-based systems. The framework exploits a multi-level hybrid Bayesian model to describe the target system and takes advantage of Bayesian inference to estimate different reliability metrics. SyRA implements several mechanisms and features to deal with the complexity of realistic models and implements a complete tool-chain that scales efficiently with the complexity of the system. The simulation time is significantly lower than micro-architecture level or RTL fault-injection experiments with an accuracy high enough to take effective design decisions. To demonstrate the capability of SyRA, we analyzed the reliability of a set of microprocessor-based systems characterized by different microprocessor architectures (i.e., Intel x86, ARM Cortex-A15, ARM Cortex-A9) running both the Linux operating system or bare metal in the presence of single bit upsets caused by radiation induced soft errors. Each system under analysis executes different software workloads both from benchmark suites and from real applications.

Path testing based reliability analysis framework of component based software system

With the increasing needs of the people in this generation, a large number of highly featured quality software systems need to be developed in all the domains. Nowadays complexity of the software system is gradually increased because of its large size. With this nature, the traditional software development process unable to produce higher quality software system within limited resources. So that, the traditional software development process has been moved to the reuse based component based software development (CBSD) which reduces the time and resource of software development. Testing is the important process in the software development life cycle to ensure the reliability or quality of software systems. Lots of reliability models have been developed to predict the software system reliability in the earlier stages of development. But these existing reliability analysis models are insufficient to estimate the reliability of component based software system (CBSS) within the limited resources. To solve this issue, the new approach was introduced by many researchers based on software architecture to estimate the reliability of component based software system. Based on that, we have proposed new framework centered on path testing to predict the reliability of the CBSS. Here we have chosen three test paths (simple, medium and complex structure) from the system for reliability estimation instead of taking all the paths. Then independent simple paths have been identified from the chosen medium and complex path to reduce the complexity of reliability estimation. All the simple paths are executed sequentially to estimate its reliability. Actual software system reliability will be predicted based on the estimated path reliability. The ATM case study has been taken to validate the proposed framework. The result obtained from this experiment is compared with the standard baseline models CUORM, LCBRM and Chao-Jung to prove the accuracy and efficiency of our proposed model. The result shows that, our proposed framework has the acceptable accuracy compared to the other models.

A general failure-pursuing sampling framework for surrogate-based reliability analysis

Design of experiment and active learning strategy are vital for the surrogate-based reliability analysis. However, the existing sampling and modeling methods usually ignore some useful information that can guide the choice of training samples, or heavily rely on the characteristics of surrogates. These lead to the inefficiency of sampling strategies or limit the application respectively. Therefore, this work proposes a failure-pursuing sampling framework, which is able to adopt various surrogate models or active learning strategies. In each iteration, it organically and sequentially takes into account the joint probability density function of random variables, the individual information at candidate points and the improvement of the accuracy of predicted failure probability. To measure the probability of the improvement, a global predicted failure probability error is proposed based on the real-time reliability analysis result. Furthermore, Voronoi diagram is applied to partition the sampling region into some local cells for keeping the uniformity of the training samples. Besides, a model-free response-distance function is developed and combined with the framework to avoid relying on the characteristics of surrogates, such as the statistical information provided by Kriging. Finally, four examples are investigated to demonstrate the applicability, stability and generality of the proposed method.

A general framework for fatigue reliability analysis of a high temperature component

AbstractFatigue reliability analysis for a high temperature component is a complex task due to the difficulties of doing experiments and uncertainties occurring in practical engineering. Researchers usually focus on the fatigue life scatter of standard specimens since the life data are easily available from fatigue tests. An alternative method is proposed in this paper which can replace the real tests of high temperature components with finite element (FE) simulation. A constitutive model coupled damage rule is utilized in this work to describe the cyclic stress‐strain response. Then, the FE simulations is incorporated into the general framework for reliability assessment. An adaptive least squares support vector machines with Latin hypercube sampling and learning function is employed to construct limit state function. To verify the accuracy and robustness of proposed method, a numerical example is utilized. In the end of this paper, a three‐layer structure extrusion container is applied for whole flowchart of fatigue reliability analysis.

SRASA: a Generalized Theoretical Framework for Security and Reliability Analysis in Computing Systems

Although there is a pressing need for highly secure and reliable computing systems, there is a glaring lack of formalism under which the properties of reliability and security can be jointly designed into these systems. This gap can primarily be attributed to the evolution of the two subfields. In the work, we introduce a unified generalized theoretical framework, called security and reliability aware state automaton (SRASA), to formally describe the specifications of a computer system that cover both security and reliability. SRASA is a 22-tuple finite state machine model that encompasses both physical and abstract states of the system, which may suffer from passive and active attacks. Three case studies illustrate the interpretation and application of the proposed SRASA theoretical framework. Our analysis and experimental results show that a non-physical attack may exploit unspecified or untested states to implement the malicious purpose, rather than introducing a new state to the system. To the best of our knowledge, this is the first attempt to bridge the current design specification gap between secure and reliable computing systems using a unified automaton approach. A general yet complete methodology that will encompass all aspects of system design, from the functional level specification to the gate level design at any system granularity, may not be feasible or it may be beyond the scope of a single work. Therefore, we aim in this work to (1) give an overview of the current landscape of reliability and security in systems design, (2) introduce a generalized framework to specify and reason about both reliability and security in the system design process, and finally (3) be general enough in the framework specification that it can be adopted or customized to more specific or concrete design instances.

A framework for model reliability and estimability analysis of crystallization processes with multi-impurity multi-dimensional population balance models

The development of reliable mathematical models for crystallization processes may be very challenging due the complexity of the underlying phenomena, the inherent Population Balance Models (PBMs) and the large number of parameters that need to be identified from experimental data. Due to the poor information content of the experiments, the structure of the model itself and correlation between model parameters, the mathematical model may contain more parameters than can be accurately and reliably identified from the available experimental data. A novel framework for parameter estimability for guaranteed optimal model reliability is proposed then validated by a complex crystallization process. The latter is described by a differential algebraic system which involves a multi-dimensional population balance model that accounts for the combined effects of different crystal growth modifiers/impurities on the crystal size and shape distribution of needle-like crystals. Two estimability methods were combined: the first is based on a sequential orthogonalization of the local sensitivity matrix and the second is Sobol, a variance-based global sensitivities technic. The framework provides a systematic way to assess the quality of two nominal sets of parameters: one obtained from prior knowledge and the second obtained by simultaneous identification using global optimization. A cut-off value was identified from an incremental least square optimization procedure for both estimability methods, providing the required optimal subset of model parameters. The implemented methodology showed that, although noisy aspect ratio data were used, the 8 most influential and least correlated parameters could be reliably identified out of twenty-three, leading to a crystallization model with enhanced prediction capability.

Reliability Modeling and Analysis of Load-Sharing Systems With Continuously Degrading Components

This paper presents a reliability modeling and analysis framework for load-sharing systems with identical components subject to continuous degradation. It is assumed that the components in the system suffer from degradation through an additive impact under increased workload caused by consecutive failures. A log-linear link function is used to describe the relationship between the degradation rate and load stress levels. By assuming that the component degradation is well modeled by a step-wise drifted Wiener process, we construct maximum likelihood estimates (MLEs) for unknown parameters and related reliability characteristics by combining analytical and numerical methods. Approximate initial guesses are proposed to lessen the computational burden in numerical estimation. The estimated distribution of MLE is given in the form of multivariate normal distribution with the aid of Fisher information. Alternative confidence intervals are provided by bootstrapping methods. A simulation study with various sample sizes and inspection intervals is presented to analyze the estimation accuracy. Finally, the proposed approach is illustrated by track degradation data from an application example.