Effect of built-in test characteristics on system reliability measures

Аuthors

Lubkov N. V.^1*, Spiridonov I. B.^2**, Stepanyants A. S.^1***

1. V.A. Trapeznikov Institute of Control Sciences of Russian Academy of Sciences, 65, Profsoyuznaya str., Moscow, 117997, Russia
2. PJSC Yakovlev , 68, Leningradskiy prospect, Moscow, 125315, Russia

*e-mail: lflbk@ipu.ru
**e-mail: igori.spiridonov@irkut.com
***e-mail: lfvrk@ipu.ru

Abstract

This article discusses the impact of built-in test dependability, fault- detection and isolation ratio on reliability of redundant systems. The results and conclusions of the article focused on responsible usage systems with high reliability and safety requirements. We present probabilistic approach to assessing the quality of built-in test based on splitting the total flow of system failures into two components — the failures, detected by the control, and latent failures. In this case the main testability index, fault detection ratio, can be defined as conditional probability of failure detection, provided that the failure occurred. The usefulness of such approach is possibility of joint reliability&testability modeling of aviation systems. We have developed a number of reliability models of typical redundant schemes that take into account parameters of built-in test. Markov reliability models were constructed for the following structures: unredundant scheme with imperfect built-in test and special recovery strategy; classical redundant structure with imperfect built-in test and graceful degradation, consisting of two parallel operating elements; redundant structure with imperfect built-in test, consisting of two parallel elements, operating in a master-slave mode; «2 out of 3» redundant scheme with repairable elements and imperfect fault coverage. Dependences of reliability and availability indices upon fault-detection ratio were investigated. Investigations were carried out on such indices as mean time to failure, probability of failure-free operation, probability of dangerous failure, operational availability factor. We have demonstrated the strong dependence of these indices on the quality of built-in test. Construction and analysis of such models allows the designer to justify the requirements to built-in test and maintenance parameters.

Keywords:

reliability, testability, redundant schemes, fault-detection ratio, fault-isolation ratio

References

1. Viktorova V.S., Vedernikov B.I., Spiridonov I.B., Stepanyants A.S. Nadezhnost’, no.3 (22), 2007, pp.62-71.

2. Tekhnicheskaya diagnostika. Terminy i opredeleniya, GOST 20911-89 (Technical diagnostics. Terms and definitions. State Standard 20911-89.) Moscow, Standarty, 1989, 8 p.

3. Tekhnicheskaya diagnostika. Kontroleprigodnost’. Obshchie trebovaniya, GOST 26656-85 (Technical diagnostics. Testability. General requirements. State Standard 26656-85), Moscow, Standarty, 1985, 15 p.

4. МIL-STD-2165. Testability program for electronic systems and equipment. Department of Defense. Washington, DC 20301, 1985.

5. Viktorova V.S., Stepanyants A.S. Modeli i metody rascheta nadezhnosti tekhnicheskikh sistem (Models and methods for calculating the reliability of technical systems), Мoscow, LENAND, 2014, 256 p.

Download