一种航空发动机性能监控的免疫神经网络模型  

Immune-neural networks model of aeroengine performance monitoring

作　　者：侯胜利[1] 王威[1] 胡金海[2] 周根娜[1] 

机构地区：[1]徐州空军学院,徐州221006 [2]空军工程大学工程学院,西安710038

出　　处：《航空动力学报》2008年第9期1748-1752,共5页Journal of Aerospace Power

摘　　要：提出了一种基于免疫神经网络的航空发动机性能监控与故障检测模型.利用人工免疫系统的识别机理,并结合人工神经网络,确定发动机性能偏离正常值的程度(异常度),实现发动机性能趋势的监控.该方法能够灵敏、准确地反映发动机整体性能的变化情况,提高发动机性能正常与否的识别率,并以此来发现发动机潜在早期故障,防止故障的扩大.通过对某型涡扇发动机进行监控,证明该方法确实有效.A model for aeroengine performance monitoring and fault detection was proposed based on immune-neural networks. With immune recognition mechanism of the artificial immune system, this model was combined with artificial neural networks to attain the degrees of engine performance＇s abnormal state for monitoring engine performance. The approach can sensitively and accurately reflect the whole performance of the engine. Meanwhile, this method improves the rate of identifying whether the performance of the engine is normal or not, finds out the potential forepart fault of engine and prevents the spread of the fault. The validity of this method is testified by monitoring certain type of turbo-fan engine.

关 键 词：免疫神经网络 航空发动机 性能监控 人工免疫系统 

分 类 号：V263.6[航空宇航科学与技术—航空宇航制造工程]