论文部分内容阅读
为有效解决某涡扇发动机过度维修造成发动机性能衰减、维修周期长、维修成本高的难题,以某涡扇发动机大修手册和维修工艺为依据,研究了基于故障检测的发动机维修流程和修理模式,建立专家诊断系统和基于BP(back propagation)神经网络的故障诊断模型,并用数台涡扇发动机真实性能数据验证故障诊断模型的可靠性,其诊断准确率高达95%,综合两者的诊断信息,制定可靠的维修方案,优化维修流程,提出了一种基于故障检测的维修决策方法.通过某涡扇真实排气温度高发动机应用验证表明:所提出的维修决策方法,有效排除故障,提高发动机的修理质量,降低维修成本,具有良好的工程应用价值.
In order to effectively solve the problem of engine performance degradation, long maintenance cycle and high maintenance cost caused by over-repair of a turbofan engine, the repair process and repair mode of the engine based on fault detection are studied on the basis of a turbofan engine overhaul manual and maintenance process, The fault diagnostic model based on BP neural network (BP) and BP neural network (BP) neural network are established. The reliability of the fault diagnosis model is verified with the real performance data of several turbofan engines. The diagnostic accuracy is as high as 95%. Based on the diagnostic information of both, To develop a reliable maintenance program and optimize the maintenance process, a maintenance decision-making method based on fault detection is proposed.According to the actual exhaust temperature of a certain turbofan, the application of the engine shows that the proposed maintenance decision-making method effectively eliminates the fault and improves the performance of the engine Repair quality, reduce maintenance costs, with good engineering value.