摘要
研究导航传感器故障诊断问题,由于飞行器导航传感器所处环境十分复杂,导航系统由多种部件组成,故障存在许多随机性、模糊性和不确定性因素,难以建立确定数学模型。传统线性模型故障诊断准确率低。为了提高飞行器导航传感器故障诊断准确率,提出一种神经网络的导航传感器故障诊断方法。飞行器导航传感器发生故障时信号中会产生突变成分,利用小波包对原始故障信号进行分解,提取信号特征向量,然后将特征向量输入神经网络训练,实现飞行器导航传感器故障智能化诊断。在Matlab平台实现传感器故障诊断的仿真,结果表明,神经网络提高了飞行器导航传感器故障诊断的准确率,是一种在线、行之有效的导航传感器故障方法。
Study navigation sensor fault diagnosis problem. Because sensors environment is very complex and faults relate each other, traditional fault diagnosis method can not diagnose fault accurately. In order to reliably find and forecast navigation sensor faults, this paper proposed a navigation sensor fault diagnosis method based on neural network. Firstly, wavelet packet was used to decompose the primitive fault signal to from signal feature vector, then feature vectors were input neural network for training, so as to realize intelligent sensor fault diagnosis. The simulation experiment was realized in Matlab platform. The simulation results show that this method increases the navigation sensor fault diagnosis accuracy and is an effective online navigation sensor fault method.
出处
《计算机仿真》
CSCD
北大核心
2012年第2期68-71,共4页
Computer Simulation
基金
数学天元基金(11026196)
关键词
小波包
神经网络
导航传感器
故障诊断
Wavelet packet
Neural network
Inertia navigation
Fault diagnosis