复杂机载电子系统故障综合诊断技术研究

故障综合诊断技术一直是复杂机载电子系统研发过程中的关键部分，当前的故障诊断技术同时需要机内测试（BIT）和场外自动化测试设备（ATE）的测试结果才能得出诊断结果，诊断效率低，时间长并且不能在线诊断。针对新一代战斗机将更加依赖航空电子系统的趋势，迫切需要一种诊断时间短，且能够实现在线诊断的故障诊断技术。因此，一种基于模型的故障诊断方法被提出。该方法通过融合多信号模型和整数编码故障字典模型，模块间采用多信号模型，单个模块中采用整数编码故障字典模型，克服了多信号模型对测试信息的浪费和整数编码故障字典模型建模困难的缺点，并提出一种多目标测试优选方法，通过优化检测方案，充分发挥BIT的检测性能。该方法通过充分使用BIT的测试信息，摆脱了对场外ATE的依赖，实现了在线快速定位故障并识别故障模式。

Research on Integrated Fault Diagnostic of Complex Avionic System

Avionic system is now a hot research topic. Integrated fault diagnostic has been a major problem of development of complex avionic system. With the increase of the dependence of aircraft on avionic system and the defects of traditional fault diagnosis methods (e.g. low diagnostic efficiency, huge diagnostic time, off-line diagnostic), this paper propose a novel diagnostic model which compromises the merits of multi-signal model and integer-coded fault dictionary. A multi-objective algorithm for test selection is also proposed in this paper. This paper has studied model-based diagnostic methods and found that a multi-objective model waste a lot of testing information, meanwhile an integer-code fault dictionary is complicated to generate, especially in large-scale systems. The proposed method overcomes these drawbacks. It allow fault diagnostic get rid of the dependence on ATE by making full use of testing information from BIT. It achieves on-line fault isolation and fault identification.