摘要
在惯性/卫星组合导航系统中,针对传统χ2检验法检测出故障但无法准确识别故障子系统的不足,提出一种基于支持向量回归的故障诊断方法.采用残差χ2检验法实时对组合导航系统进行故障检测,并构建基于支持向量机的回归预测模型,实现对惯性导航系统状态的预测;根据系统模型输出和预测模型输出之差辅助进行惯性导航系统的故障判别,诊断出系统故障源.仿真结果表明,所提出的方法能够快速准确地识别故障子系统,并进行有效的系统隔离和重构,从而使组合导航系统的性能得到保障.
Aiming at the problem that the traditional residual chi-square test can detect the fault but can＇t identify the fault subsystem in an integrated GPS/Inertial navigation system, a fault diagnosis method based on support vector regression is proposed. The residual chi-square test is used to do real-time detection for the integrated navigation system. The support vector regression prediction model is constructed to realize the prediction of the inertial navigation system. The output difference between the system model and the prediction model is an auxiliary to diagnose the system fault source. The simulation results show that the method can fast and accurately identify the fault subsystem, and thus fault isolation and system reconfiguration can be done to ensure the performance of the integrated navigation system.
出处
《控制与决策》
EI
CSCD
北大核心
2016年第10期1889-1893,共5页
Control and Decision
基金
国家自然科学基金项目(61174197
61428303)
关键词
支持向量回归
组合导航
惯性导航系统
卫星导航系统
故障检测
support vectorregression
integrated navigation
inertial navigationsystem
global positioningsystem
faultdetection
