论文部分内容阅读
为了提高航空发动机故障检测正确率,将BP神经网络应用于航空发动机故障检测中。从某航空公司使用的CFM56-7B系列发动机的实际飞行历史数据中选取研究样本,对比了6种训练方法的效果并最终选择弹性BP法对网络加以训练并进行测试。结果表明:该方法对CFM56-7B系列发动机的排气温度指示故障、进口总温指示故障和可调放气活门故障的检测正确率高达83.33%。BP神经网络能够很好地应用于航空发动机的实际故障检测,其学习记忆稳定、网络收敛速度快,具有一定的工程实用价值。
In order to improve the correctness of aero-engine fault detection, BP neural network is applied to aero-engine fault detection. From the actual flight history data of a CFM56-7B series engine used by an airline, a sample of the study is selected, the effectiveness of the six training methods is compared, and finally the elastic BP method is selected to train and test the network. The results show that this method can detect the exhaust gas temperature of the CFM56-7B series engine, and the correctness of the total inlet temperature indication fault and the adjustable bleed valve fault is as high as 83.33%. BP neural network can be well applied to the actual fault detection of aeroengine, which has the advantages of stable learning and memory, fast network convergence, and certain practical engineering value.