Reliability prediction of imperfect switching systems subject to multiple stresses

Abstract

Reliability prediction plays a very important role in system design, and the two key factors considered in predicting system reliability are: failure distribution of the component/equipment and system configuration. This paper discusses about the imperfect switching system with one component in active and k spares in the standby state. When the operating component breaks down, the switch will be able to detect the failure using the sensor and replace the defective component with a functionable spare, so the system can keep operating. Therefore, the switch and the sensor have direct impact on normal operations of the switching systems. The reliability of two types of imperfect switching system is thoroughly discussed and compared: (1) a non-repairable system with only one standby component, one switch and one sensor and (2) a non-repairable system with two standby components, one switch and one sensor. Since gamma distribution is fairly adequate to describe the failure mechanism of a system under k-times of shock multiple stresses. This paper then assumes in system (1) and (2), the operating components follow gamma failures, sensor and switch failures follow exponential distribution. In addition, three modes are assumed in regards to the switch failure: under energized, under failing-open and under failing-closed condition. This paper uses MAPLE computer language to perform reliability estimation and comparison on the above-mentioned systems with components, switch and sensor under different failure rate and various intended period of use. Its results can provide guidelines on decision making for improving system design in industries.