基于PCA与DBN的航空发动机气路系统故障诊断
|
摘要
提出了一种基于主元分析与深度信念网络相结合的航空发动机气路系统故障诊断新方法。首先利用主元分析方法对发动机监测参数原始数据进行降维,以便降低参数数据的复杂度,获得较低维数的最优特征参数。其次,采用深度信念网络对获取的最优特征参数数据创建故障诊断模型。为验证所创建诊断模型的准确性,采用某型航空发动机气路系统实测数据对所创建的基于主元分析与深度信念网络故障诊断模型进行实验验证,结果表明,经过主元分析降维后的深度信念网络故障诊断模型的诊断结果比未经过主元分析降维后的深度学习神经网络故障诊断模型的诊断结果准确率更高,从而验证了该方法的有效性。
This paper proposes a new method for fault diagnosis of aero-engine gas pass system based on a combination of principal component analysis(PCA)and deep belief network(DBN).First,in order to reduce the complexity and get the optimal feature parameters,the PCA method is used to reduce the dimension of the original engine monitoring data.Second,a DBN neural network is used to create a fault diagnosis model for the obtained optimal feature parameters.To confirm the accuracy,an experimental verification of the fault diagnosis model based on the PCA and DBN is conducted by using the measured data of an aero-engine gas path system.The results show that the diagnostic accuracy of the DBN neural network fault diagnosis model with the PCA is better than that of the model without it.This verifies the effectiveness of the proposed method.
This paper proposes a new method for fault diagnosis of aero-engine gas pass system based on a combination of principal component analysis(PCA)and deep belief network(DBN).First,in order to reduce the complexity and get the optimal feature parameters,the PCA method is used to reduce the dimension of the original engine monitoring data.Second,a DBN neural network is used to create a fault diagnosis model for the obtained optimal feature parameters.To confirm the accuracy,an experimental verification of the fault diagnosis model based on the PCA and DBN is conducted by using the measured data of an aero-engine gas path system.The results show that the diagnostic accuracy of the DBN neural network fault diagnosis model with the PCA is better than that of the model without it.This verifies the effectiveness of the proposed method.
引文
[1] YAHUI WU.Aeroengine rub-impact fault diagnosis based on wavelet packet transform and the local discriminate bases[J].Applied Mechanics and Materials Vols,2012,226-228:740-744.
[2] 刘远航,王铁楠,郑丽.航空发动机故障在线监测系统的研究[J].中国新技术新产品,2018(9):49-50.
[3] MARIA GRAZIA DE GIORGI.A dignostics tool for aero-engines health monitring using machine learing technique[J].Energy Procedia,2018,148:860-867.
[4] G?3MEZ,M.J.Automatic condition monitoring system for crack detection in rotating machinery[J].Reliability Engineering & System Safety,2016,152:239-247.
[5] 李本威,林学森,杨欣毅,等.深度置信网络在发动机气路部件性能衰退故障诊断中的应用研究[J].推进技术,2016,37(11):2173-2180.
[6] GUOXING LAN.Comparison and fusion of various classification methods applied to aero-engine fault diagnosis[C].东北大学、IEEE新加坡工业电子分会、中国自动化学会信息物理系统控制与决策专业委员会,沈阳,2017.
[7] 段隽喆,李华聪.基于故障树的故障诊断专家系统研究[J].科学技术与工程,2009,9(7):1914-1917.
[8] 费成巍,艾延廷,王蕾,等.基于支持向量机的航空发动机整机振动故障诊断技术研究[J].沈阳航空工业学院学报,2010,27(2):29-32+19.
[9] 吴月伟,杨慧.基于RBF神经网络的飞机发动机故障诊断研究[J].计算机工程,2005(S1):266-267.
[10] 车畅畅,王华伟,倪晓梅,等.基于深度学习的航空发动机故障融合诊断[J].北京航空航天大学学报,2018,44(3):621-628.
[11] 崔建国,吴灿,董世良,等.主元分析法和模糊积分的航空发动机气路状态监测[J].火力与指挥控制,2014,39(10):130-134.
[12] 龚志飞,郭迎清.基于主元分析法的航空发动机传感器故障诊断研究[J].计算机测量与控制,2012,20(08):2017-2019+2023.
[13] 张琰,李忠海,蒋丽英.基于改进主元分析的故障诊断方法研究[J].沈阳航空工业学院学报,2009,26(2):54-57.
[14] 李本威,林学森,杨欣毅,等.深度置信网络在发动机气路部件性能衰退故障诊断中的应用研究[J].推进技术,2016,37(11):2173-2180.
[15] 单晨晨.基于深度置信网络的发动机状态监控研究[D].天津:中国民航大学,2017.
[16] 石鑫,朱永利.深度学习神经网络在电力变压器故障诊断中的应用[J].电力建设,2015,36(12):116—122.
[17] 彭军,郭晨阳,张勇,等.基于深度学习的航空发动机部件故障诊断[J].系统仿真技术,2018,14(01):20-24.
[2] 刘远航,王铁楠,郑丽.航空发动机故障在线监测系统的研究[J].中国新技术新产品,2018(9):49-50.
[3] MARIA GRAZIA DE GIORGI.A dignostics tool for aero-engines health monitring using machine learing technique[J].Energy Procedia,2018,148:860-867.
[4] G?3MEZ,M.J.Automatic condition monitoring system for crack detection in rotating machinery[J].Reliability Engineering & System Safety,2016,152:239-247.
[5] 李本威,林学森,杨欣毅,等.深度置信网络在发动机气路部件性能衰退故障诊断中的应用研究[J].推进技术,2016,37(11):2173-2180.
[6] GUOXING LAN.Comparison and fusion of various classification methods applied to aero-engine fault diagnosis[C].东北大学、IEEE新加坡工业电子分会、中国自动化学会信息物理系统控制与决策专业委员会,沈阳,2017.
[7] 段隽喆,李华聪.基于故障树的故障诊断专家系统研究[J].科学技术与工程,2009,9(7):1914-1917.
[8] 费成巍,艾延廷,王蕾,等.基于支持向量机的航空发动机整机振动故障诊断技术研究[J].沈阳航空工业学院学报,2010,27(2):29-32+19.
[9] 吴月伟,杨慧.基于RBF神经网络的飞机发动机故障诊断研究[J].计算机工程,2005(S1):266-267.
[10] 车畅畅,王华伟,倪晓梅,等.基于深度学习的航空发动机故障融合诊断[J].北京航空航天大学学报,2018,44(3):621-628.
[11] 崔建国,吴灿,董世良,等.主元分析法和模糊积分的航空发动机气路状态监测[J].火力与指挥控制,2014,39(10):130-134.
[12] 龚志飞,郭迎清.基于主元分析法的航空发动机传感器故障诊断研究[J].计算机测量与控制,2012,20(08):2017-2019+2023.
[13] 张琰,李忠海,蒋丽英.基于改进主元分析的故障诊断方法研究[J].沈阳航空工业学院学报,2009,26(2):54-57.
[14] 李本威,林学森,杨欣毅,等.深度置信网络在发动机气路部件性能衰退故障诊断中的应用研究[J].推进技术,2016,37(11):2173-2180.
[15] 单晨晨.基于深度置信网络的发动机状态监控研究[D].天津:中国民航大学,2017.
[16] 石鑫,朱永利.深度学习神经网络在电力变压器故障诊断中的应用[J].电力建设,2015,36(12):116—122.
[17] 彭军,郭晨阳,张勇,等.基于深度学习的航空发动机部件故障诊断[J].系统仿真技术,2018,14(01):20-24.