飞行器电源系统在线实时健康管理

来源期刊:中南大学学报(自然科学版)2014年第6期

论文作者:冯威 于劲松 姜杨 刘忠

文章页码:1867 - 1873

关键词:电源系统;健康管理;编译贝叶斯网络;在线实时

Key words:electrical power system; health management; compiling Bayesian network; online real-time

摘    要:飞行器电源系统在线实时健康管理需解决观测信息不确定性、任务完成时限性的问题,针对此问题,总结国内外最新健康管理算法,研究贝叶斯网络、编译贝叶斯网络,提出飞行器电源系统在线实时健康管理方法。首先基于系统结构模型,采用面向对象方法构建贝叶斯网络健康模型,然后通过全局消元算法对构建的贝叶斯网络健康模型进行离线编译,所得到的运算电路健康模型则可在线实时计算系统健康状态的后验概率。实验结果表明:与贝叶斯网络健康模型相比,运算电路健康模型在观测信息不确定性的情况下,不仅能够高精度在线实时诊断飞行器电源系统故障,也可以有效满足其健康管理严格时限性要求。

Abstract: To address the problem of observational information uncertainties and task deadlines met in the online real-time health management for aerial electrical power system, the latest HM algorithms in China and aboard were summarized which were focused on the Bayesian networks (BNs) and compiling Bayesian network, and then an approach of online real-time health management for aerial electrical power system was proposed. In this approach, the object-oriented method was used to develop BN health model based on the system structure model, and then the BN was offline compiled into the arithmetic circuit (AC) health model using global variable elimination algorithm, subsequently, the resulting circuit can be applied to calculate the posterior probability of system health. The experimental results show that the AC health model, compared with the BN health model, not only precisely online real-time isolates the system faults, but also meets the strict time deadlines of HM under the condition of observational information uncertainties.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号