集成时间信息的液体火箭发动机故障诊断推理

Fault diagnosis reasoning for liquid-propellant rocket engines with integration of time information

提出一种液体火箭发动机集成时间信息的定性建模和故障诊断推理方法。该方法以谓词逻辑公式和子句的形式对发动机定性特征和试车过程中的观测信息加以描述,同时基于归结原理和假言推理规则的演绎推理方法进行诊断问题求解。结合实际试车数据的检验结果表明,该方法为液体火箭发动机试车过程中动态诊断知识的表达和组织,提供了一种简洁有效的方式,计算量小,检测速度快,且具有较强的诊断能力。

A qualitative modeling and diagnosis method was proposed with the integration of time information for the fault diagnosis of liquid-propellant rocket engines. The modeling method is based on the qualitative description of observation and characteristics of engines in the form of predication logic formulae and clauses, and the diagnosis method is based on the inference rules of resolution step and modus ponens. Results verified with the on-fire test data show that the method proposed provides an efficient way it is easy to represent the dynamic diagnostic knowledge of the test processes. The proposed method needs less computation and has better ability in the fault detection and diagnosis in contrast to normal ones.