航空发电机综合故障诊断技术研究

Comprehensive fault diagnose technology research of aircraft generator

针对航空发电机的故障难以准确诊断的问题,提出了一种基于BP神经网络的航空发电机综合故障诊断方法,通过对BP神经网络数学原理及算法的深入分析与研究,构建基于BP神经网络的航空发电机故障诊断模型。结合某型航空发电机的真实试验数据,对所构建故障诊断模型的正确性进行了试验验证。在试验验证过程中,利用航空发电机真实试验数据的前60组对神经网络进行训练,后40组数据对神经网络进行测试,最终验证了所构建的故障诊断模型能够较好实现对航空发电机故障进行准确诊断的效能,且所采用的BP神经网络方法具有收敛速度快、识别能力强、精度高以及准确性高等优点。

According to the problem which aircraft generator fault is difficult to be accurately diagnosed,it puts forward a aircraft generator fault diagnosis methods based on BP neural network.The paper analyses in-depth and researches the mathematical principle for BP neural network and algorithm,and constructs aircraft generator fault diagnosis model based on BP neural network.The paper combines aircraft generator real test data to verify the correctness of the aircraft generator fault diagnosis model.In the experimental verification process,it uses the first 60 groups of real test data of aircraft generator to train the neural network,and uses the last 40 sets of data to test the neural network.Finally,it turns out that the aircraft generator fault diagnosis model can be better realized the aircraft generator comprehensive fault diagnose efficiency,and the BP neural network method has the advantage of quickly convergence speed,good recognition ability and the high accuracy.