论文部分内容阅读
基于故障树模型的诊断把故障树的底事件分成三部分：必然的故障源集（ＣＦＳ）、正常底事件集（ＮＥＳ）和可能的故障源集（ＰＦＳ），及如何进一步确定ＰＦＳ中各元素的状态（正常或异常）。在存在大量训练样本的情况下，可采用基于神经网络模型的学习诊断方法来确定ＰＦＳ中各元素的状态，并通过对某卫星能源系统故障模拟原理性试验台的故障诊断验证了该方法的有效性。
Based on the fault tree model diagnosis, the bottom of the fault tree is divided into three parts: the necessary fault source set (CFS), the normal bottom event set (NES) and the possible fault source set (PFS), and how to further determine the elements in the PFS State (normal or abnormal). In the presence of a large number of training samples, the learning diagnosis method based on neural network model can be used to determine the state of each element in the PFS, and the effectiveness of the method is verified through the fault diagnosis of a simulation test bed of a satellite energy system fault Sex.