Reliability Engineering最新文献

Global reliability (g-reliability) of any complex system is defined and two simplified different methods for evaluating the same are proposed. Both the proposed methods are conceptually simpler and computationally efficient for g-reliability evaluation of any large complex system and effectively applicable to its proper form. They therefore require less computer memory and computational efforts as compared to other existing methods. Proposed methods are applicable to the network having both nodes and branches of finite non-zero failure probability. The efficiency of each method increases with the complexity of the network. Both the proposed methods are also illustrated with examples to demonstrate the effectiveness of each method.

In order to establish a feasible and useful reliability evaluation model for a network system of dependent components, the failure of a component is distinguished into many states artificially, according to the causes which bring about such a failure and the effects to the system in such a way that the failure rate of each state can be easily estimated. Such failure state of a component can be grouped into four types, single (stochastical independent), active, passive and common-cause failures.

One of the main influences on the reliability of sodium sulphur cells has been identified to be the cracking of beta alumina ceramic. Several ceramic degradation mechanisms have been proposed; flaw-dependent, subcritical crack growth, stress corrosion cracking and progressive degradation mechanisms. These are reviewed in the paper. Reliability models based on these mechanisms are considered, and methods to distinguish between the models are developed. None of the models seems clearly to account for the empirical results obtained. The authors develop an alternative model. The method developed may be used for other similar processes.

Let F be a life distribution with survival function $\text{F}\text{= 1 −}\text{F}$ and finite mean $\text{μ = ∫}{}_0{}^{\mathrm{\infty }}\text{F}\text{(}\text{x}\text{)d}\text{x}$. The scaled total time on test transform is defined by . In this paper, the properties of $\text{φ}{}_{\mathrm{F}}\text{(t)}$ for some criteria of the mean residual life are investigated. These criteria include decreasing mean residual life average, decreasing harmonic mean residual life average, new better than used harmonic mean residual and new better than used mean residual life. Some test statistics for testing exponentially against all these mean residual life criteria are proposed. The distributions of the test statistics are investigated for small statistics samples. Powers of some of these tests are estimated by simulation.

We consider a system which supports a load in such a way that the profit earned by the system for one time unit is a continuous and increasing function of the amount of the load. The time-to-failure distribution of the system is assumed to be gamma. Both parameters of the distribution are continuous functions of the load such that the mean time-to-failure is a decreasing function of the load.

The system earns the profit only when it is on. When it is off, it is repaired, and the losses are assumed to be proportional to the repair time. We define the average asymptotic gain earned by the system in a single time unit; this quantity can be thought of as a system efficiency measure.

Assuming that the load is a controllable variable, the goal is to find a value of the load that maximizes the so-defined efficiency. To model the relationship between the load and the parameters of the time-to-failure distribution (gamma) as well as the relation between the load and the profit, power and exponential functions have been taken.

The analytical conditions for the existence of maximum positive gain and the analysis of the behaviour of the gain as the load varies are presented.

The Availability of a system, device or component in an operational cycle, defined by normal operation—malfunction (failure) — wait — repair, is considered and defined in the form of a random variable. By the use of the fast convolution techniques originally developed for structural reliability applications, the cumulative distribution function of this random variable is obtained together with its mean (equivalent to the long run Average Availability) and standard deviation. The technique described includes preventive maintenance, may be used with any assumed probability distributions of failure, waiting, repair and preventive maintenance times, and may take account of correlation between these parameters. The results are equivalent to those which may be obtained by conventional numerical simulation of repair cycles, but the procedure described is likely to be faster and more economical of computing time and amenable to use on microcomputers.

A programme of operator error analysis is being carried out for the Sizewell ‘B’ PWR design. This paper describes the methods used, the analysis which has been carried out, preliminary results and design changes which have been incorporated to reduce the risk. At this stage, systematic identification and reduction of risk from operator error has been achieved. Further analysis and quantification of error is required to substantiate that the risk arising from operator error does not dominate the risk due to mechanical plant failures.