An online reliability evaluation method based on evidential reasoning for space relay

Current reliability assessment approaches targeted at dynamic systems suffer from the challenge of uncertainties. This paper, we proposed a new online reliability evaluation method which based on evidential reasoning, which can integrate both historic and present status information to online assess the reliability of space relay under uncertainties. Firstly, the space relay is analyzed and the characteristic parameters which can reflect the health status are selected as reliability data. Secondly, the parameter data are collected and updated online weights using the variation-based weighting method which are further normalized. Thirdly, the appropriate forgetting factors are determined with the consideration of the influences on the assessment result from both historic and present data in the system operation process. Finally, a fusion scheme based evidential reasoning algorithm is applied to integrate the historic and present reliability data which are online monitored, and the system's present health status is then assessed. A case study for space relay reliability evaluation is examined to corroborate the effectiveness of the presented method. The result shows that the reliability of the space relay can be precisely evaluated and the presented method has also proved superior property in assessing the health state of dynamic systems.