为了提高航天装备面对突发灾害和战术任务的实时响应性，星载综合电子设备作为整星的核心管控设备亟需提高在轨快速自测试能力。因此本项目提出将航空与武器电子领域的BIT(Built-in-test,BIT)设计理念引入星载综合电子设备，然而传统BIT存在虚警率高、故障诊断能力弱等致命缺点，不能满足高可靠星载综合电子设备的需求。为了解决该问题，开展面向星载综合电子设备的智能BIT关键技术研究。首先通过某型号星载综合电子设备的故障模式、机理与影响分析，确定设备级的特征参数和时间应力参数；其次，提出采用特征级信息融合的故障检测算法，综合考虑多特征参数的变化趋势特征、统计特征、多参数之间的关联冗余特征、时间应力信息的评估等；最后，采用故障Petri网进行故障建模和推理，针对故障Petri网对故障传播过程及时间特征建模能力不足的缺点，提出了一种改进的故障Petri网模型和故障诊断推理算法。

In order to improve the real-time rapid response of aerospace equipment in the face of sudden disaster and tactical mission, as the core management and control equipment of the satellite, the onboard integrated avionics need to improve the rapid self test ability. So the project imports the BIT (Built-in-test, BIT) design concept of aviation and weapon electronics into the onboard integrated aerospace avionics, however, the traditional BIT has fatal disadvantages, such as high false alarm rate and weak fault diagnosis, which can not meet the demand of high reliability of onboard integrated avionics. So as to solve the problem, the research on the key technologies of intelligent BIT for aerospace avionics is carried out. First of all, through the analysis of the failure mode, mechanism and effect of aerospace avionics, the product characteristic parameters and relevant time stress parameters will be determined; Secondly, the fault detection algorithm based on feature level information fusion is proposed, in the method, the parameters’ statistical characteristics, change trend time stress assessment information and redundant mathematical relationship among multi-parameters are comprehensively considered. Finally, fault modeling and reasoning based on fault Petri net is employed, due to the insufficient modeling ability of fault broadcast and time-related diagnosis knowledge, an improved fault Petri net model and fault diagnostic reasoning algorithm is put forward.