基于代价敏感的序贯三支决策最优粒度选择方法

最优粒度选择是序贯三支决策领域研究的热点之一,旨在通过合理的粒度选择来对复杂问题进行求解。在现阶段最优粒度选择中,代价敏感是影响决策的重要因素之一。针对这个问题,该文首先基于信息增益和卡方检验提出一种新的属性重要度计算方法;其次,为了更好地符合实际应用场景,在构建多粒度空间时将代价参数与粒度大小相结合,设置了相应的惩罚规则,并分析了决策阈值的变化规律;最后,为了消除测试代价和决策代价量纲不一致所带来的影响,借助变异系数设计了一种客观的代价计算方法。实验结果表明,该模型适用于现有代价认知场景,能在给定代价情况下选出代价最小的最优粒层。     

基于代价敏感的序贯三支决策最优粒度选择方法

最优粒度选择是序贯三支决策领域研究的热点之一,旨在通过合理的粒度选择来对复杂问题进行求解.在现阶段最优粒度选择中,代价敏感是影响决策的重要因素之一.针对这个问题,该文首先基于信息增益和卡方检验提出一种新的属性重要度计算方法;其次,为了更好地符合实际应用场景,在构建多粒度空间时将代价参数与粒度大小相结合,设置了相应的惩罚规则,并分析了决策阈值的变化规律;最后,为了消除测试代价和决策代价量纲不一致所带来的影响,借助变异系数设计了一种客观的代价计算方法.实验结果表明,该模型适用于现有代价认知场景,能在给定代价情况下选出代价最小的最优粒层.     

Grouped Genetic Algorithm Based Optimal Tests Selection for System with Multiple Operation Modes

To insure various operation modes operated properly, important modules or potential fault sources must be monitored. Due to the hardware resource limit, only part of modules and potential faults can be covered by online test. The percentage of modules or fault sources been covered is termed as fault detection rate (FDR). Lower mode FDR constraints must be satisfied to insure the whole mission executed successfully. It’s a typical multi-constraint optimization problem. In this paper, a grouped genetic algorithm (GGA) is proposed to minimize the test cost and satisfy the FDR constraints. Each GA is used to optimize the test set for one mode. The final chromosome of each GA represents the selected tests in one mode. The union set of the chromosomes of all GAs gives the final solution. Each GA is subjected to one FDR constraint, therefore, the optimal solution is more likely be found. The group of GAs are executed in parallel, hence, the proposed method is efficient. The effectiveness and efficiency of the proposed method are verified by statistical experiments.     

测试不可靠条件下基于量子进化算法的测试优化选择

测试优化选择是复杂电子系统测试性设计中的一个重要问题.首先从测试容差的角度分析了测试发生漏检和虚警的原因,在此基础上建立了测试不可靠条件下一种新的测试选择模型,模型以测试代价、漏检代价和虚警代价之和最小为优化目标,以故障检测率和故障隔离率为约束条件;然后提出一种改进的量子进化算法对模型求解,该算法通过改进一种已有可靠测试选择算法而成,包括种群初始化、适应度计算和种群的进化策略.最后通过两个仿真实例验证了求解算法及模型的有效性和优越性.     

Sequential Testing of Components for System Reliability

It is often desirable to construct s-confidence limits for system reliability on the basis of data obtained from `pass-fail' tests on the components of the system. This paper presents a general method for sequentially testing the components that provides data from which these s-confidence limits can be easily derived. The method is applicable to any s-coherent system for which the reliability function is known. It is a generalization of a scheme given by Winterbottom and Verrall for systems composed of units arranged either in series or parallel.     

厚膜金导体的可靠性试验

介绍了厚膜金导体可靠性试验方面的一些研究结果。对金的电化学迁移以及厚膜金导体的热试验进行了分析 ,并讨论了在工艺过程中应注意的一些问题。     

A Technique for Optimal Sequential Maintenance Scheduling

This paper presents the mathematical development of, and a simple solution technique for, an optimal sequential maintenance scheduling problem. The model is shown to satisfy the Kuhn-Tucker conditions, the necessary and sufficient conditions for the existence of an optimal solution. Real-world systems consisting of mixtures of constant and increasing failure rate devices are considered in the model. Sequential preventive maintenance schedules are developed for groups of identical items with increasing failure rates. Provision is made for the corrective maintenance of these groups if failures occur in between the preventive maintenance schedules. Also, constant failure rate devices are accorded corrective maintenance when failures occur. Optimality is achieved by minimizing the total annual maintenance cost, subject to constraints on the system availability, number of maintenance personnel and intervals of preventive maintenance. The model is applied to a coal mine power system example.     

Computer Usage Monitoring for Design and Reliability Tests

The usage conditions of a product or system can be monitored and analyzed to provide information to improve the design and qualification of future products. This paper discusses a method to collect and analyze product usage conditions and use this data for product design and testing. A case study of the usage monitoring of commercial computers is presented. Three parameters - temperature, humidity, and vibration - were monitored inside computers being operated by a variety of users. The collected data was compared with the product specifications. Four group tests were conducted. The first group test simulated different human usage conditions, such as lifting the computer, inserting it into a bag, dropping it, and so on. The second group test involved daily usage monitoring on different users (mainly office and home users). The third group test involved usage monitoring of a single user during different time periods. The fourth group test was an experiment carried out for validation tests in the lab. The relationship between the real-life load levels and product qualification test load levels was then established.     

Truncated Sequential Life Tests for a 3-Way Decision Procedure

Eleven sets of test plans are given for testing ``? = ?0' against the alternative ``? = ?1 < ?0' and ``? = ?2 > ?0' where ? is the failure rate. The results are useful for incentive plans and life-cycle costing applications, and are fundamental in the life testing of equipment, whether a consumer or industrial item, during the constant failure-rate period. The theory and equations for the calculations of the tables and plans are developed and discussed and the related references are included.     

光通信有源器件的可靠性试验

本文主要报导日本光纤通信用光发射和探测器件的可靠性试验。     

Confidence Limits for System Reliability: A Sequential Method

A sequential method is given for obtaining confidence limits for system reliability when ``pass-fail' test data have been obtained on components. Costs of testing are examined and a rule is derived for determining the order in which to test the components so that the average cost of testing is minimized. The method is presented first for series systems, but in Section IV the extension to parallel systems is considered and in Section V the procedure is applied to a series-parallel configuration. Comparisons are given with fixed sample-size testing and it is shown that the sequential method is statistically efficient and that many of the difficulties encountered when using fixed sample sizes are eliminated by testing sequentially.     

Sequential Destruction Method for Monte Carlo Evaluation of System Reliability

Circumstances favoring the use of Monte Carlo methods for evaluating the reliability of large systems are discussed. A new method, that of Sequential Destruction (SD) is introduced. The SD method requires no preparatory topological analysis of the system, and remains viable when element failure probabilities are small. It applies to a variety of reliability measures and does not require element failures to be s-independent. The method can be used to improve the performance of selective sampling techniques. Substantial variance reductions, as well as computational savings, are demonstrated using a sample system with more than 100 elements.     

Optimal Signaling and Selection Verification for Single Transmit-Antenna Selection

In marked contrast with the ideal error-free feedback assumption that is common in the literature, practical systems are likely to have severely bandwidth-limited, error-prone feedback channels. We consider the scenario where feedback from the receiver is used by the transmitter to select the best antenna, out of many available antennas, for data transmission. Feedback errors cause the transmitter to select an antenna different from the one signaled by the receiver. We show that optimizing the signaling assignment, which maps the antenna indices to the feedback codewords, improves performance without introducing any additional redundancy. For a system that uses error-prone feedback to transmit quadrature-phase-shift-keying-modulated data from a single antenna selected from many available spatially correlated antennas, we derive closed-form approximations for the data symbol error probability for an arbitrary number of receive antennas. We use these to systematically find the optimal signaling assignments using a low-complexity algorithm. The optimal signaling is intimately coupled to how the receiver performs selection verification, i.e., how it decodes the data signal when, due to feedback errors, it does not always know which antenna was used for data transmission. We show that ignoring feedback errors at the receiver can lead to an unacceptable performance degradation, and develop optimal and suboptimal, blind and nonblind selection-verification methods. With a small side-information overhead, nonblind verification approaches the ideal perfect selection-verification performance     

Optimal Design for System Reliability and Maintainability

A complex system consisting of N modules that are logically interconnected for mission success is to be placed in the field for a fixed period of time. For some modules, standby units can be provided, for others this option is not available, but we must select from among several design alternatives differing in cost, weight, and reliability. The problem is to determine simultaneously the module designs and the numbers of standby units to maximize the system reliability, subject to cost and/or weight constraints. Other authors have considered a similar problem for a pure series inter-connection of modules, but we permit the system to be any configuration of modules in series and/or parallel. A dynamic programming model is presented for this problem. The notion of the generalized decomposition operator is used to develop a set of recursive relations. An example is included.     

Optimal Relay Selection for Energy-Efficient Multicast

We consider multicast transmission from a single source to multiple destinations. We assume that the source cannot reach the destinations directly, but must forward its traffic through a set of assisting relay nodes. The performance objective under consideration is to maximize the common amount of information (number of bits) that the source delivers to all destinations per joule of the total energy spent. Our aim is to obtain a policy that identifies: (a) which subset of the relays should be activated, (b) for how long, and (c) the respective destinations that each relay has to serve. We consider centralized policies with exact knowledge of the channel conditions. In the special case of networks employing at most two relays, we show that for any fixed assignment of destinations to relays the problem of maximizing the number of bits per joule by choosing the duration that each relay should be activated can be formulated as a convex optimization problem. Unfortunately, the problem of assigning destinations to relays is combinatorially complex. Thus, in the sequel we present a method with reduced complexity that exploits the knowledge of the underlying channel conditions to perform this assignment. Finally, we provide a set of numerical results to illustrate the optimal relay selection and assignment of destinations to relays corresponding to different channel conditions.     

基于加速退化试验数据的可靠性分析

本文研究了基于加速退化试验数据的可靠性评估技术,给出了加速退化方程和加速退化因子的定义并结合实例研究了加速退化失效模型的可靠性统计推断方法.在工程问题中使用加速退化试验将更能节省时间和费用.     

Combining GAs and Symbolic Methods for High Quality Tests of Sequential Circuits

A symbolic fault simulator is integrated in a Genetic Algorithm (GA) environment to perform Automatic Test Pattern Generation (ATPG) for synchronous sequential circuits. In a two phase algorithm test length and fault coverage as well are optimized. Furthermore, not only the Single Observation Time Test Strategy is supported, but also test patterns with respect to the Multiple Observation Time Test Strategy are generated. However, there are circuits that are hard to test using random pattern sequences, even if these sequences are genetically optimized. Thus, deterministic aspects are included in the GA environment to improve fault coverage. Experiments demonstrate that both a priori time consuming strategies, the symbolic simulation approach and the GA, can be combined at reasonable costs: Tests with higher fault coverages and considerably shorter test sequences than previously presented approaches are obtained.     

广播电视系统及设备可靠性与环境测试

本文介绍了广播电视系统及设备进行可靠性与环境测试的原则和方法,重点介绍了一些主要在室外使用的设备在测试时所参照的标准和测试方法。此外,介绍了应用于广播电视系统及设备的可靠性与环境测试中所使用的主要设备仪器。以上测试的开展,为广播电视系统和设备的稳定性、可靠性测试提供了实践基础,保证了整个广播电视设备制造行业的良性的、可持续发展。     

序贯Bayes概率比检验在雷达可靠性试验中的应用

依据经典概率统计原理设计试验方案进行舰炮雷达靶场可靠性试验,试验统计分析需要较大样本量,实施难度大。利用Bayes序贯概率比检验,可以有效解决这一问题。本文从经典概率统计可靠性检验入手,详细介绍基于Bayes序贯概率比检验理论的可靠性检验原理及方案,并结合某型雷达试验实例进行分析,介绍其在雷达可靠性试验中的应用。     

序贯Bayes概率比检验在雷达可靠性试验中的应用

依据经典概率统计原理设计试验方案进行舰炮雷达靶场可靠性试验,试验统计分析需要较大样本量,实施难度大。利用Bayes序贯概率比检验,可以有效解决这一问题。本文从经典概率统计可靠性检验入手,详细介绍基于Bayes序贯概率比检验理论的可靠性检验原理及方案,并结合某型雷达试验实例进行分析,介绍其在雷达可靠性试验中的应用。