基于粗糙神经网络的航空电子传感器故障诊断

针对航空电子传感器系统是由多个子系统所组成的复杂大系统且很难量测系统输入,传统故障诊断方法难以评估这类系统的问题,提出了一种基于粗糙神经网络的航空传感器故障诊断方法。该方法首先运用Kohonen网络对航空电子传感器测量得到的连续数据进行离散化,然后用粗糙集理论进行知识规则的提取,最后用提取的知识作用于一步预测神经网络,用该网络预测结果与航空电子传感器实际输出进行阈值比较,进而进行故障检测。仿真实验和实际应用表明,该方法可行并能有效地检测传感器故障,且故障诊断率高。

Avionics sensor fault diagnosis based on rough set and neural network

The avionics sensor system is a complex system.It is difficult to diagnose avionics sensor fault by using traditional method.Diagnostic method based on neural network and rough set is presented.The data of avionics sensor system used by Kohonen network is discretizes.It applies the rough set theory to discover knowledge of discrete measure data.It builds the neural network to forecast the sensor output data.It gives the diagnose result by analyzing the neural network output and avionics sensor real-time output.The simulation and application show that the method is feasible and available.