Vibration prediction method of aircraft external stores based on deep belief network

被引：0
|
作者：
Liu Z. [1 ]
Xu J. [1 ]
Hu X. [2 ]
机构：
[1] Products Department of Equipment Service, China Aero-polytechnology Establishment, Aviation Industry Corporation of China, Limited, Beijing
[2] Naval Equipment Department, the Chinese People's Liberation Army, Beijing
来源：
Hangkong Dongli Xuebao/Journal of Aerospace Power | 2021年 / 36卷 / 06期
关键词：
Aircraft external stores; Deep belief network; Environmental worthiness; Model optimization; Vibration prediction;
D O I：
10.13224/j.cnki.jasp.2021.06.008
中图分类号：
学科分类号：
摘要：
In view of the non-linear relationship between the vibration environment, altitude and Mach number of the aircraft external stores in flight, a method for predicting the vibration environment of aircraft external stores with deep belief network (DBN) was proposed. GJB 150.16A empirical formulas were analyzed and then a total of 7470 sets of simulation data were generated, by which linear regression, polynomial regression and DBN models were set up respectively, and DBN number of hidden layer node was optimized, then the expected errors of three methods, including root mean square error, mean absolute error and mean relative error, were compared and analyzed. DBN method was applied to a case for predicting the vibration of a certain aircraft store, aiming to verify the feasibility and accuracy of DBN method. The results showed that the DBN model can fully characterize the non-linear relationship between the pressure altitude, Mach number and RMS value of the aircraft external stores in the flight environment, and the overall expected effect was better than linear regression and polynomial regression. The expected mean relative error was about 2.24dB, and reduced by more than 50%. This method has provided a new way to predict the vibration environment of aircraft external stores, and is of great significance for prognostics health management in the whole life of aviation weapon equipment. © 2021, Editorial Department of Journal of Aerospace Power. All right reserved.
引用
收藏
页码：1197 / 1205
页数：8
相关论文
共 20 条
  • [1] HINTON G, OSINDERO S, THE Y., A fast learning algorithm for deep belief nets, Neural Computation, 18, 7, pp. 1527-1554, (2006)
  • [2] TAMILSELVAN P, WANG Pingfeng, Failure diagnosis using deep belief learning based health state classification, Reliability Engineering and System Safety, 115, pp. 124-135, (2013)
  • [3] TRAN V T, ALTHOBIANI F, BALL A., An approach to fault diagnosis of reciprocating compressor valves using Teager-Kaiser energy operator and deep belief networks, Expert Systems with Applications, 41, 9, pp. 4113-4122, (2014)
  • [4] ZHANG Guohui, Research on time series prediction and its application based on deep belief network, (2017)
  • [5] WANG Shuqin, Gas time series prediction and anomaly detection based on deep learning, (2018)
  • [6] HE Jun, Vibration characteristic analysis and intelligent fault diagnosis of gearboxes, (2018)
  • [7] LANGKVIST M, KARLSSON L, LOUTFI A., Sleep stage classification using unsupervised feature learning, Advances in Artificial Neural Systems, 2012, 5, (2012)
  • [8] HINTON G, SALAKHUTDINOV R., Reducing the dimensionality of data with neutral networks, Science, 313, 5786, pp. 504-507, (2006)
  • [9] HINTON G., A practical guide to training restricted Boltzmann machines, pp. 599-619, (2012)
  • [10] WANG Guanglu, XU Ming, LI Dapeng, Aircraft flight vibration prediction methodology, Advances in Aeronautical Science and Engineering, 1, 3, pp. 251-255, (2010)
12