论文部分内容阅读
卫星动量轮在轨运行环境复杂,并且无失效数据,难以利用传统方法进行可靠性建模与评估。为此,提出一种利用贝叶斯网络融合动量轮各种试验信息及轴温和电流等遥测数据的可靠性建模与评估方法。基于失效分析建立贝叶斯网络拓扑结构,根据试验数据估计网络参数,而后,通过贝叶斯网络的推理得到动量轮可靠度的点估计和区间估计,并利用在轨遥测数据实时评估动量轮可靠性,获得动量轮可靠性变化趋势比较明显的遥测数据取值区间。该方法对动量轮实时状态监控具有一定的理论和实践意义。
Satellite momentum wheel in-orbit operating environment is complex, and no failure data, it is difficult to use traditional methods for reliability modeling and evaluation. For this reason, a method of reliability modeling and evaluation based on Bayesian network fusion of various test information and shaft temperature and current telemetry data is proposed. Based on the failure analysis, the Bayesian network topology is built, the network parameters are estimated based on the experimental data, and then the point estimates and interval estimates of the momentum rounds are obtained by Bayesian network reasoning. , Get the range of telemetering data with the obvious trend of momentum reliability change. The method has certain theoretical and practical significance to the real-time monitoring of momentum wheel.