论文部分内容阅读
针对部队复杂系统故障诊断中存在的诊断精度低,虚警率高等问题,提出一种测试不可靠条件下多故障诊断方法.为解决系统诊断贝叶斯网络结构和概率映射表建立困难的问题,通过建立系统的多信号流图模型,从而获得系统诊断贝叶斯网络.将测试不可靠度引入概率映射表,增加了算法工程应用中的鲁棒性.利用后验概率诊断推理将问题归结为不等式约束极值问题,采用0-1规划隐数法对不等式极值问题求解,从而获得最优解.以某型导弹制导系统电子部件为例,验证了该方法对复杂系统多故障诊断的有效性.
Aiming at the problems of low diagnostic accuracy and high false alarm rate in the fault diagnosis of complex military systems, a multi-fault diagnosis method based on unreliable test conditions is proposed.In order to solve the problem of system diagnosis of Bayesian network structure and probability mapping table, By establishing a multi-signal flow graph model of the system, a system diagnosis Bayesian network is obtained, which introduces the test unreliability into the probability mapping table and increases the robustness in the application of algorithm engineering. The problem can be attributed to Inequality constrained extreme value problem, the 0-1 planning hidden number method is used to solve the extreme inequality problem to obtain the optimal solution. Taking an electronic component of a missile guidance system as an example, the method is proved to be effective for multi-fault diagnosis of complex systems Sex.