基于涡扇发动机模型的数据生成及应用方法

【作者】 王鹏； 郭迎清； 孙浩； 毛皓天；

【机构】 西北工业大学动力与能源学院；

【摘要】 为了进一步的解决在航空发动机故障诊断算法研究过程中,在故障开发阶段存在缺少原始数据的问题。对发动机数据的生成与发动机模型数据的生成进行了较为深刻的研究。以民用大涵道比涡扇发动机模型为对象可以生成该发动机的所有数据,以发动机闭环模型为对象可以生成发动机模型的所有数据。之后便可以根据发动机故障仿真及诊断算法的验证测试对发动机数据的要求,来提供不同的数据,让这两个过程能够在不同的数据下进行,而不是继续单单的局限于某一组特定的数据。这样做能为这两个过程提供充足的数据源从而让这两个过程在进行的时候有更多的选择,从而可以用评价准则对算法做出评判,找到能够让这两个过程进行的较好的一组或几组发动机数据。最终实现发动机故障数据仿真更真实,算法验证测试更准确的目的,从而较好的解决上述问题。更多还原

【Abstract】 In order to further solve the problem that the original fault data is missing in the fault development stage in the research process of the engine fault diagnosis algorithm, the engine data generation and engine model data generation have been studied in depth. All data of the engine can be generated using the civil large bypass ratio turbofan engine model, and all engine model data can be generated with the engine closed loop model. Afterwards, that can provide different data according to the requirements of the validation of the engine fault simulation and the test of diagnosis algorithm, so that the two processes can be performed under different data condition, instead of continuing to be limited to only one group of specific data condition. Doing so can provide sufficient data sources for these two processes so that the two processes can have more choices at the time of the process, so that the evaluation criteria can be used to judge the algorithm and find out in which of one group or few groups specific data that the two processes can perform better. Finally, the engine fault data simulation is more realistic, and the algorithm verification test is more accurate, so as to better solve the above problems.更多还原