以串聯觀點探討系統不同失效模式之效應

In series conception to study the effect on distinc failure mode of system

呂紹澤(Shao-ze Lu)

指導教授 ： 王國雄

國立中央大學/工學院/機械工程學系/碩士(2012年)

摘要

本論文係以串聯觀點，探討系統不同失效模式之效應，目的希望藉由失效誘因觀點，評估系統內子系統誘使系統失效之機率與可靠度退化過程。本文首先由變異係數建立潛伏性指標，藉以將系統失效之潛伏性質與以量化，並針對韋伯分佈模型與代數型模型之潛伏性質與反曲點進行探討，結果發現上述二種模型各自之形狀參數與其潛伏性指標和反曲點對應可靠度，彼此呈現一對一對一關係。

對於串聯形式可靠度模型以及h(R)模型，皆可透過相關之轉換得到類似混合形式可靠度模型之累加性質，如此可對具累加性質之串聯系統成份與以分類，藉以評估子系統誘使系統失效之機率，以及子系統壽命與可靠度退化過程等資訊。

對於部分串聯系統之子系統間通常彼此具有互動性，但傳統對於串聯形式可靠度模型，係基於子系統間彼此互斥或獨立之假設條件而成立，如此對於子系統彼此之互動關係則無法有效呈現，因此，本論文以h(R)模型建立可描述串聯系統子系統可靠度關係式，子系統彼此亦具有如串聯系統之正交性質，而無須基於子系統間彼此需互斥或獨立之假設條件，可將子系統間之互動性與以呈現。

關鍵字

可靠度 ； 串聯系統 ； 潛伏性指標 ； 失效誘因

並列摘要

This thesis is based on the series conception to study the system in distinct failure mode of the effect, for the purpose by conception of failure incentive to estimate that system within the subsystems to induce system failure of probability and reliability degradation process.

First of this article established latent indicators (or called cumulative index) by the coefficient of variation, in order to quantify the latent nature of the system, to aimed at the Algebraic-Equation model and the Weibull model of the latent nature and inflection point to explore, and found each of the two model that the shape parameters and latent indicators and Rinf, showing the one-on-one relationship.

Series form of the reliability model and h(R) form model, which can be through related conversion similar to the mixed form of reliability model of the added nature, hence can classification the components of series system of these added nature, to evaluate the induce probability of system to failure and MTTF and reliability of degradation processes of the subsystems.

The subsystems of part of the series systems usually possess of interactive with each other, but the traditional series form of the reliability model, which mutually exclusive or independent of assumptions based on the subsystem set up, so that cannot effectively show the relationship between the interaction of the subsystems to each other. Therefore, in this thesis purposed that h(R) model to describe the subsystem reliability relationship and the orthogonal nature between each other in series systems, instead of needed to base on the subsystem between mutually exclusive or independent assumption conditions, it is the interaction between subsystems to present.

並列關鍵字

reliability ； series system ； cumulative index

參考文獻

[2]. Paul Kales, Reliability : for technology, engineering, and management

,Prentice-Hall, New Jersey, 1998.

[3]. Dimitri B. Kececioglu, Reliability & Life Testing Handbook, Volume 1, DEStech, U.S.A., 1993.

[4]. U Dinesh Kumar, John Crocker, T Chitra, Haritha Saranga, Reliability and Six Sigma, Springer Science, New York, 2006.

[6]. M. Xie and C.D. Lai, “Reliability analysis using an additive Weibull model with bathtub-shaped failure rate function”, Reliability Engineering and System Safety, Vol 52, pp.87-89, 1995.