摘要
以神经网络的自学习代替故障诊断观测器法中的解析计算，提出了一种新的故障诊断方法，从而解决了具有模型不确定性系统的故障诊断问题。对某型航空发动机控制系统传感器故障进行了仿真研究，该方法不仅具有传统观测器法报警及时准确、易于实现容错控制的优点，又不需要获得系统精确的数学模型。
An Artificial Neural Network(ANN)-based observer failure detection method is presented,which uses ANN's self-learning ability instead of analytical computation of the obsever design to get the redundancy relationship of the system.In order to deal with the failure detection of model uncertainty system,the failure detection of the control system sensor of a given aeroengine has been simulated with the presented method.The simulation shows that this method not only has the advantage of the observer failure detection method,but also has no need for the precise model of system.It is useful to the failure detection of the model uncertainty system.
出处
《航空动力学报》
EI
CAS
CSCD
北大核心
1997年第2期149-151,共3页
Journal of Aerospace Power
