基于神经网络的红外型空空导弹故障诊断专家系统

普通的导弹自动测试系统只能给出主要参数的检测结果,针对以上不足,通过对神经网络和专家系统在自动测试领域中的深入研究,设计完成了一种基于神经网络的故障诊断专家系统,克服了传统专家系统故障定位慢、容易出错、数据库难修改的缺点,提高了装备保障工作的时效性和可靠性。重点阐述了知识库和推理机的构建思路和方法,采用VC＋＋6.0进行软件设计。通过大量实验验证得出,该系统执行速度快、故障定位准确,具有广阔的应用前景。

Fault Diagnosis Expert System for Infrared Air-to-Air Missile Based on Neural Network

Common missile automatic test systems can merely provide the test results of main parameters.To resolve this problem,a new fault diagnosis expert system based on the neural network is designed in this paper.The shortcomings of conventional expert systems,such as being slow to locate the fault,easy to make mistakes and hard to revise database,have been overcomed,which improves the timeliness and reliability of equipment support work.The ideas and means of building the knowledge base and reasoning machine are clarified in detail.VC ＋ ＋6.0 is adopted to design the software program.It has been proved through abundant experiments that the system performs promptly,locates the fault accurately and owns a wide application prospect.