航空发动机滑油系统建模及测试性分析

Modeling and Testability Analysis of Aviation Engine Oil System

摘要:

针对日常维修中滑油系统不易到达且重要度高的特点,开展基于仿真模型的滑油系统测试性分析,优化滑油系统诊断策略。基于全流量单回路反向循环滑油系统的工作原理,使用AMESim软件建立滑油系统的虚拟仿真模型,并利用Simulink模型仿真结果验证所建AMESim模型的准确性,误差在3.5%以内；建立了滑油系统基本故障事件的贝叶斯网络,分析滑油系统的五大故障模式,用于确定滑油系统重要故障事件,作为开展仿真模型故障注入研究的基础；在模型中实现故障注入,通过分析预置测试点参数的变化,得到故障-测试相关性矩阵,实现滑油系统测试性分析和优化,优化后故障检测率提高了16.7%,故障隔离率提高了40.0%。可以利用测试性分析的结果,优化滑油系统传感器布局设计,为滑油系统的故障诊断提供支撑。

Abstract:

In order to address the characteristics of the oil system that is difficult to reach and has high importance in daily maintenance, a test analysis of the oil system based on a simulation model was conducted to optimize the diagnostic strategy of the oil system. Based on the working principle of the full-flow single-loop reverse circulation oil system, a virtual simulation model of the oil system was established using AMESim software, and the accuracy of the built AMESim model was verified by using Simulink model simulation results, with an error within 3.5%. A Bayesian network of basic fault events of the oil system was established to analyze the important fault modes of the oil system, which was used to determine important fault events of the oil system as a basis for conducting fault injection research based on simulation models. Fault injection was implemented in the model, and by analyzing the changes in preset test point parameters, a fault-test correlation matrix was obtained to achieve testability analysis and optimization of the oil system. After optimization, the fault detection rate increased by 16.7% and the fault isolation rate increased by 40.0%. The results of testability analysis can be used to optimize the sensor layout design of the oil system and provide support for fault diagnosis of the oil system.