Exact equation and an algorithm for reliability evaluation of K-out-of-N:G system

An expression for reliability of K-out-of-N:G system is proposed. An algorithm for computing reliability of K-out-of-N system is given. It is an easy to implement, fast and memory efficient algorithm and helps to improve the computational efficiency considerably.     

Reliability measurement of mass storage system for onboard instrumentation

Advances in spaceborne vehicular technology have made possible the long-life duration of the mission in harsh cosmic environments. Reliability and data integrity are the commonly emphasized requirements of spaceborne solid-state mass storage systems, because faults due to the harsh cosmic environments, such as extreme radiation, can be experienced throughout the mission. Acceptable dependability for these instruments has been achieved by using redundancy and repair. Reconfiguration (repair) of memory arrays using spare memory lines is the most common technique for reliability enhancement of memories with faults. Faulty cells in memory arrays are known to show spatial locality. This physical phenomenon is referred to as fault clustering . This paper initially investigates a quadrat-based fault model for memory arrays under clustered faults to establish a reliable foundation of measurement. Then, lifelong dependability of a fault-tolerant spaceborne memory system with hierarchical active redundancy, which consists of spare columns in each memory module and redundant memory modules, is measured in terms of the reliability (i.e., the conditional probability that the system performs correctly throughout the mission) and mean-time-to-failure (i.e., the expected time that a system will operate before it fails). Finally, minimal column redundancy search technique for the fault-tolerant memory system is proposed and verified through a series of parametric simulations. Thereby, design and fabrication of cost-effective and highly reliable fault-tolerant onboard mass storage system can be realized for dependable instrumentation.     

Software dependability models under memory faults with application to a digital system in nuclear power plants

In this work, software dependability under memory faults in the operational phase is predicted by two models: an analytic model and the stochastic activity network (SAN) model. The analytic model is based on the simple reliability theory and the graph theory, which represents the software as a graph composed of nodes and arcs. Through proper transformation, the graph can be reduced to a simple two-node graph from which software reliability can be derived. The SAN model permits the representation of concurrency, timeliness, fault tolerance, and degradable performance of the system and provides a means for determining the stochastic behavior of a software.Using these models, we predict the reliability of an application software in a digital system, Interposing Logic System (ILS), in a nuclear power plant and show the sensitivity of software reliability to major physical parameters which affect software failure in the normal operation phase. It is found that the effects of hardware faults on software failure should be considered for the accurate prediction of software dependability in the operation phase.     

Discussion of compensating faults in modelling of NMR system reliability

To obtain realistic estimates of reliability of systems with N-tuple Modular Redundancy (NMR) the effect of compensation of logical faults should be considered. Throughout the analyses made up to date, different models of compensating faults have been considered, producing results that are rather impractical and yield very complex mathematical formulas for reliability indices. In this paper a general model is presented that combines the ideas most commonly applied. As a main result some useful estimates are given of upper and lower bounds of reliability indices of NMR systems with compensating faults.     

Reliability analysis of K-out-of-N: G systems with dependent failures and imperfect coverage

This paper presents a Markov model for reliability analysis of K-out-of-N: G systems subject to dependent failures with imperfect coverage. Closed form solutions of the probabilities are used to obtain the reliability and the mean time to failure (MTTF). A numerical example is provided to illustrate the results.     

Concurrent error correction in unidirectional linear arrays

Abstract

A concurrent error detection and correction algorithm for errors caused by permanent, intermittent and transient faults in a unidirectional linear array is introduced. The fault model we assumed here is based on some unknown reasons and may affect a small area of the IC chip. The structure based on this time‐redundant technique is proposed. The simplicity of the adding circuitry to a normal linear array makes this structure very attractive in VLSI implementation. The error‐correction capability and the total processing time for an n operands k stages processes are also discussed.     

Fault‐tolerant procedures for redundant computer systems

Real‐time computer systems deployed in life‐critical control applications must be designed to meet stringent reliability specifications. The minimum acceptable degree of reliability for systems of this type is ‘7 nines’, which is not generally achieved. This paper aims at contributing to the achievement of that degree of reliability. To this end, this paper proposes a classification scheme of the fault‐tolerant procedures for redundant computer systems (RCSs). The proposed classification scheme is developed on the basis of the number of counteracted fault types. Table I is created to relate the characteristics of the RCSs to the characteristics of the fault‐tolerant procedures. A selection algorithm is proposed, which allows designers to select the optimal type of fault‐tolerant procedures according to the system characteristics and capabilities. The fault‐tolerant procedure, which is selected by this algorithm, provides the required degree of reliability for a given RCS. According to the proposed graphical model only a part of the fault‐tolerant procedure is executed depending on the absence or presence (type and sort) of faults. The proposed methods allow designers to counteract Byzantine and non‐Byzantine fault types during degradation of RCSs from N to 3, and only the non‐Byzantine fault type during degradation from 3 to 1 with optimal checkpoint time period. Copyright © 2008 John Wiley & Sons, Ltd.     

Design fault tolerance

Typical software fault tolerance techniques are modeled on successful hardware fault tolerance techniques. The software fault tolerance techniques rely on design redundancy to tolerate residual design faults in the software; the hardware fault tolerance techniques rely on component redundancy to tolerate physical degradation in the hardware. Investigations of design redundant software have revealed difficulties in adapting the hardware strategy to software.We survey three categories of issues: (1) practical issues in the implementation of design-redundant software, (2) economic considerations for the development and maintenance of multiple software implementations, and (3) assessment difficulties in measuring and predicting the performance of design-redundant software. All of these issues should be considered by would- be developers of design-redundant software to justify use of the technique.     

A parallel built-in self-diagnostic method for nontraditional faults of embedded memory arrays

In this paper, we propose a built-in self-diagnostic march-based algorithm that identifies faulty memory cells based on a recently introduced nontraditional fault model. It is developed based on the DiagRSMarch algorithm, which is a diagnostic algorithm to identify traditional faults for embedded memory arrays. A minimal set of additional operations is added to DiagRSMarch for identifying the nontraditional faults without affecting the diagnostic coverage of the traditional faults. The embedded memory arrays are accessed using a bidirectional serial interfacing architecture which minimizes the routing overhead introduced by the diagnosis hardware. Using the concepts of the bidirectional interfacing technique, parallel testing, and redundant-tolerant operations, the diagnostic process can be accomplished efficiently at-speed with minimal hardware overhead.     

FATIGUE CRACK GROWTH UNDER MIXED MODES I AND II

Fatigue crack growth behaviour under mixed modes I and II was studied by applying in-phase alternating tensile and torsional loading to a thin-walled hollow cylindrical specimen with an initial crack. In the linear region of a log-log plot where da/dN=A(ΔK)^m, da/dN at first decreases with increasing ΔK₁₁₀ component and then approaches a minimum close to the value of ΔK₁₁₀/ΔK₁₀～ 0.58; here ΔK₁₁₀/ΔK₁₀ is the ratio of the initial ΔK_II to the initial ΔK₁., When ΔK₁₁₀/ΔK₁₀ increases further, da/dN increases. Under shear mode, da/dN becomes higher than that under mode I. The ΔK₁, and ΔK₁₁ components during fatigue crack growth under mixed mode loading increase and decrease, respectively, with an increase in da/dN In the low crack growth rate region the fatigue crack growth rates accelerate with an increase of the initial ΔK₁₁ component, ΔK₁₁₀. Fatigue life increases with increase of ΔK₁₁₀/ΔK₁₀ under the test condition of equivalent stress range being kept constant and the pre-crack length being the same.     

Multi-objective genetic algorithm for solving N-version program design problem

N-version programming (NVP) is a programming approach for constructing fault tolerant software systems. Generally, an optimization model utilized in NVP selects the optimal set of versions for each module to maximize the system reliability and to constrain the total cost to remain within a given budget. In such a model, while the number of versions included in the obtained solution is generally reduced, the budget restriction may be so rigid that it may fail to find the optimal solution. In order to ameliorate this problem, this paper proposes a novel bi-objective optimization model that maximizes the system reliability and minimizes the system total cost for designing N-version software systems. When solving multi-objective optimization problem, it is crucial to find Pareto solutions. It is, however, not easy to obtain them. In this paper, we propose a novel bi-objective optimization model that obtains many Pareto solutions efficiently.We formulate the optimal design problem of NVP as a bi-objective 0–1 nonlinear integer programming problem. In order to overcome this problem, we propose a Multi-objective genetic algorithm (MOGA), which is a powerful, though time-consuming, method to solve multi-objective optimization problems. When implementing genetic algorithm (GA), the use of an appropriate genetic representation scheme is one of the most important issues to obtain good performance. We employ random-key representation in our MOGA to find many Pareto solutions spaced as evenly as possible along the Pareto frontier. To pursue improve further performance, we introduce elitism, the Pareto-insertion and the Pareto-deletion operations based on distance between Pareto solutions in the selection process.The proposed MOGA obtains many Pareto solutions along the Pareto frontier evenly. The user of the MOGA can select the best compromise solution among the candidates by controlling the balance between the system reliability and the total cost.     

一种基于Triplet loss的齿轮箱复合故障识别方法

随着设备检测点的数量与采样频率的增加,机械健康监测进入了"大数据"时代。深度学习以其强大的自适应特征提取和分类能力也在机械大数据处理方面取得了丰硕的成果。在故障诊断领域,目前深度学习方法的研究对象均集中于单一故障,而复合故障却鲜有人涉足。复合故障因为其各类故障信号间有耦合,变化的工况(负载,转速)也会对信号产生较大影响,所以难以准确诊断。面对复杂的复合故障,传统的Softmax分类器已不能精确高效的完成故障诊断。提出了一种基于Triplet loss的深度度量学习模型的诊断方法,对齿轮箱的轴承及齿轮这两种目标的故障同时进行诊断。其优势在于通过该模型提取故障信号的特征,再利用Triplet loss度量各类故障之间的距离,使得同类故障特征间的距离很近,异类故障特征间的距离很远,从而高效完成诊断任务。试验结果表明,该方法实现了在多种工况,大量样本下对齿轮箱内轴承和齿轮不同故障的准确诊断。     

Transient analysis of reliability with and without repair for K-out-of-N:G systems with two failure modes

This paper presents Markov models for transient analysis of reliability with and without repair for K-out-of-N:G systems subject to two failure modes. The reliability of repairable systems can be calculated as a result of the numerical solution of a simultaneous set of linear differential equations. Closed form solutions of the transient probabilities are used to obtain the reliability for nonrepairable systems.     

Tables of Tolerance-Limit Factors for Normal Distributions

Tables of factors for use in computing two-sided tolerance limits are presented. In contrast to previous tabulations of the tolerance-limit factor K, we tabulate the factors r(N, P) and u(f, γ), whose product is equal to K. This results in greatly increased compactness and flexibility. The mathematical development is discussed, including methods used to compute the tabulated values and a study of the accuracy of the basic approximation. A number of possible applications are discussed and examples given.     

Transient analysis of reliability with and without repair for K-out-of-N :G systems with M failure modes

This paper represents Markov models for transient analysis of reliability with and without repair for K-out-of-N :G systems subject to M failure modes. The reliability and the mean time between failures of repairable systems can be calculated as a result of numerical solution of simultaneous set of linear differential equations. Closed form solutions of the transient probabilities are used to obtain the reliability and the mean time to failure for nonrepairable systems.     

Defect and Transient Fault-Tolerant System Design for Hybrid CMOS/Nanodevice Digital Memories

Targeting on the future fault-prone hybrid CMOS/nanodevice digital memories, this paper presents two fault-tolerance design approaches that integrally address the tolerance for defects and transient faults. These two approaches share several key features, including the use of a group of Bose-Chaudhuri-Hocquenghem (BCH) codes for both defect tolerance and transient fault tolerance, and integration of BCH code selection and dynamic logical-to-physical address mapping. The first approach is straightforward and easy to implement but suffers from a rapid drop of achievable storage capacity as defect densities and/or transient fault rates increase, while the second approach can achieve much higher storage capacity under high defect densities and/or transient fault rates at the cost of higher implementation complexity and longer memory access latency. Based on extensive computer simulations and BCH decoder circuit design, we have demonstrated the effectiveness of the presented approaches under a wide range of defect densities and transient fault rates, while taking into account of the fault-tolerance storage overhead and BCH decoder implementation cost in CMOS domain     

Effective K-nearest neighbor classifications for Wisconsin breast cancer data sets

ABSTRACT

Using machine learning algorithms for early prediction of the signs and symptoms of breast cancer is in demand nowadays. One of these algorithms is the K-nearest neighbor (KNN), which uses a technique for measuring the distance among data. The performance of KNN depends on the number of neighboring elements known as the K value. This study involves the exploration of KNN performance by using various distance functions and K values to find an effective KNN. Wisconsin breast cancer (WBC) and Wisconsin diagnostic breast cancer (WDBC) datasets from the UC Irvine machine learning repository were used as our main data sources. Experiments with each dataset were composed of three iterations. The first iteration of the experiment was without feature selection. The second one was the L1-norm based selection from the model, which used the linear support vector classifier feature selection, and the third iteration was with Chi-square-based feature selection. Numerous evaluation metrics like accuracy, receiver operating characteristic (ROC) curve with the area under curve (AUC) and sensitivity, etc., were used for the assessment of the implemented techniques. The results indicated that the technique involving the Chi-square-based feature selection achieved the highest accuracy with the Canberra or Manhattan distance functions for both datasets. The optimal K values for these distance functions ranged from 1 to 9. This study indicated that with the appropriate selection of the K value and a distance function in KNN, the Chi-square-based feature selection for the WBC datasets gives the highest accuracy rate as compared with the existing models.

Abbreviations: KNN: K-nearest neighbor; Chi2: Chi-square; WBC: Wisconsin breast cancer     

Analysis of crack propagation using ΔK and Kmax

In this investigation a general relationship between fatigue crack growth rate, da/dN, and a two-parameter ΔK and K_max driving force is derived using fundamental fatigue (ε–N curve) properties. A power-law relationship between ΔK and K_max is obtained by relating the crack growth rate to the fatigue life of the ‘process zone’. Theoretically, there are four different regions on a log–log plot depending on the particular combinations of ΔK and K_max. The actual analysis of experimental data indicates only two different regions namely, ΔK and K_max dominated, corresponding to high and low load ratios, respectively. A new way of representing the da/dN data in terms of ΔK and K_max by means of the crack propagation (CP) table is proposed. Finally, the application of the CP table for predicting crack growth rate under constant amplitude loading is explained and discussed.     

Parameters affecting the damage tolerance behaviour of railway axles

The paper provides a discussion on damage tolerance options applied to railway axles and factors influencing the residual lifetime as well as the required inspection interval. These comprise material properties such as the scatter of the da/dN–ΔK curve, the fatigue crack propagation threshold ΔK_th and the toughness of the material. Parameters affecting axle loading such as the press fit, rotating bending, load history and mixed crack opening modes are discussed. Finally the influence of the initial crack geometry on residual lifetime is simulated.