Research On Reliability Modeling And Analysis Of Artificial Satellite System

Artificial Satellite is belong to a long life and high reliability product, so the failure sample size is extremely limited. In order to make use of the limited artificial satellite failure samples to guide the reliability design and test of the artificial satellite system, it is essential to seek effective and applicable reliability modeling analysis of statistical method in the case of small samples.As the non-parametric method can reflect the real information of the original sample. In this paper, the reliability analysis method of the small samples about complex system has been studied, and the Johnson mean rank order method and the Kaplan-Meier method are used to model reliability of the spacecraft system, and the double non-parametric bootstrap method is employed to evaluate the artificial satellite system.The parametric method is very flexible,and it is widely used in reliability engineering. In this paper, as the sectional Weibull model can fit various forms of data samples, it is used to model artificial satellite system. The specific modeling analysis process: first, analyze the failure rate of reliability of artificial satellite system, and the number of sectional model. Subsequently, use Weibull transform to change reliability and fit it in WPP. Then, use the least squares method to estimate parameters sectional Weibull model. Finally, analyze the fitting results, and use double non-parametric bootstrap method to evaluate the applicability of the model.