基于人工智能方法的传感器故障诊断技术

A fault detection method based on artificial intelligence for sensors

在液体火箭发动机地面试验过程中,传感器的失效率远高于发动机部件、组件的故障率,因而提出了采用基于人工智能的方法对传感器的故障进行检测与诊断。本文对基于人工智能方法用于传感器故障检测与诊断的特点进行了分析,并分别应用BP神经网络和自适应神经模糊推理系统(ANFIS)对某次传感器的故障进行检测。结果证明基于人工智能的方法稳定可靠,具有良好的工程应用前景。

In rocket engine ground hot tests, the probability of sensor fault is much larger than components of the rocket engine, so methods based on artificial intelligence （AI） are proposed for sensor fault detection and diagnosis. The characteristics of AI methods are analyzed and a factual example is presented by Back Propagation neural network and Adaptive Neural Fuzzy Inference System （ANFIS）. The result shows that the AI methods are steady in running, so it is of great engineering application value.