Research On Reliability Modeling And Evaluation Of Satellite Attitude Control System

The development of satellite technology is a comprehensive reflection of the national space level, Satellite reliability analysis and life prediction is one of the key technologies in the field of space. In the complex space environment, the interaction between the ray and the material produced by the electric ion causes the ionization, excitation, nuclear reaction and chemical reaction of the internal atoms of the material. In addition, there are many factors, such as mechanical wear, electronic products and the attenuation of optical devices, which results in the performance degradation and failure of satellite components and devices. Therefore, it is of great significance to study the performance degradation of the key components of the satellite and predict the lifetime of the satellite.The satellite momentum wheel is the key component of the satellite attitude and orbit control system, and its failure or failure will directly cause the satellite to lose the ability to adjust the attitude according to the instruction. In order to ensure the successful completion of the mission, the momentum wheel is usually required for a few years or even ten years of time to work steadily. However, because of the limitation of technology, fund and time, the traditional reliability engineering technology can not satisfy the reliability test of momentum wheel. Therefore, the actual project of on orbit satellite in close contact, study new momentum wheel reliability modeling and life prediction method, improve the assessment accuracy, is the urgent problem of China's satellite momentum wheel in the process of development. The fiber optic gyroscope as another important key components of the satellite, its reliability has a direct impact on the life of the satellite, the performance is also significant degradation of reliability analysis and life prediction data of fiber optic gyro.The accurate prediction of the remaining life of the equipment is the basis of health management, and the key problem of the engineering practice. For this kind of complex equipment, it is difficult to model the evolution law of random variation, but the traditional method based on life data is difficult to implement for small sample and high cost equipment. Therefore, based on the state monitoring data for the degradation modeling and residual life prediction, and then to achieve the management decision-making technology has become the current research frontier in the field of reliability engineering. However, there are still a lot of theoretical difficulties related to engineering practice.First of all, this paper summarizes the research methods, the performance degradation of reliability analysis and life prediction based on data analysis secondly, failure mechanism of the momentum wheel, using the degradation data of momentum wheel is established, the life prediction and reliability analysis of linear Wiener process and gamma process. Then, the degradation mechanism of fiber optic gyro is analyzed, and the degradation model of nonlinear Wiener process is established by using the degradation data. At the same time, several new degradation models are proposed and simulated. Finally, using Bayesian updating online life prediction for a single product, but also on the degradation data with measurement error is processed using Kalman filtering principle, estimate the hidden state, the parameters are updated online, the prediction effect is very good.