论文部分内容阅读
对独立电源系统进行故障诊断与预测研究是保证整个复杂运动装置系统安全性工程的重要环节.分析了目前独立电源故障诊断系统中存在的问题,提出采用多传感器信息融合和多智能体技术相结合的方法来提高故障诊断的可靠性和系统的扩展性.利用智能体的自主性、分布性和协作性,构建了独立电源多智能体信息融合故障诊断系统.根据独立电源故障征兆的特点,将D-S(Dempster-Shafer)证据理论引入到多神经网络的诊断结果融合技术中,阐述了多神经网络局部诊断智能体和D-S证据理论融合诊断智能体的具体实现方法.最后,以某型航空电源故障诊断为例,给出了故障实例的诊断仿真,结果表明该方法可有效提高诊断可信度.
The research on fault diagnosis and prediction of independent power system is an important part of the security engineering of the entire complex motion system.Analyze the existing problems in the independent power fault diagnosis system and propose a multi-sensor information fusion and multi-agent technology Method to improve the reliability of fault diagnosis and the scalability of the system.According to the characteristics of independent power failure symptoms, the independent power supply multi-agent information fusion fault diagnosis system is constructed based on the autonomy, distribution and cooperation of agents, DS (Dempster-Shafer) theory of evidence is introduced into the fusion of diagnostic results of multi-neural networks, and the concrete realization methods of multi-neural network local diagnostic agent and DS evidence theory fusion diagnostic agent are described.Finally, with a certain type of aviation power failure As an example, the diagnostic simulation of fault cases is given. The results show that this method can effectively improve the diagnostic credibility.