|Table of Contents|

Improved BP Neural Network System for Engine Performance Trend Analysis and Fault Diagnosis

《南京理工大学学报》（自然科学版）[ISSN:1005-9830/CN:32-1397/N]

Issue:
2010年01期
Page:
24-29
Research Field:
Publishing date:

Info

Title:
Improved BP Neural Network System for Engine Performance Trend Analysis and Fault Diagnosis
Author(s):
LIU Yong-jian1; ZHU Jian-ying1; ZENG Jie2
1.College of Civil Aviation;2.College of Aerospace Engineering,Nanjing University ofAeronautics and Astronautics,Nanjing 210016,China
Keywords:
neural network;  ant colony optimization algorithm;  Levenberg-Marquardt algorithm;  aeroengines;  performance trend analysis;  fault diagnosis
PACS:
V263.6;TP183
DOI:
-
Abstract:
Aiming at the disadvantage of conventional BP neural network,such as selecting parameter values by the empirical method,slow convergence speed and easy trap into local minimum points,this paper designs an improved BP neural network system.In order to improve the network convergence rate and reduce the training error,this paper optimizes the initial connecting weight value of neural network by the use of ant colony algorithm and trains artificial neural network by Levenberg-Marquardt(LM) algorithm.The algorithm is applied to an engine for access to the aircraft performance parameters of aeroengine trend analysis and fault diagnosis by using ACARS messages in real time.It can analyse the aeroengine performance trend and diagnose complex fault quickly and accurately.Finally,the improved algorithm is tested through simulation and the results show that the improved algorithm can get higher confidence level than the previous algorithm.

References:

[ 1] H ag anM T, Demuth H B, BealeM H. 神经网络设计[M ]. 戴葵, 等译. 北京: 机械工业出版社, 2002:45- 78.
[ 2] 宋红英. 基于蚁群神经网络的发动机故障诊断专家系统的研究[ D ]. 北京: 中国农业大学图书馆,2005: 36- 43.
[ 3] 冯今朝, 王仲生. 基于LM 优化算法的神经网络在航空发动机转子故障诊断中的应用[ J] . 宇航计测技术, 2007( 4): 18- 21.[ 4 ] 姚华, 单贵平, 孙健国. 基于卡尔曼滤波器及神经网络的发动机故障诊断[ J]. 航空动力学报, 2008,23( 6): 1112- 1117.
[ 5] M ishra D, Y adav A, Ray S, et a .l Levenberg-M a rquardtlearn ing a lgo rithm for Integrate-and-F ireNeuron-M ode l[ J] . N eural In fo rm ation Processing-Letters andRev iew s, 2005, 9( 2): 45- 53.
[ 6] 余宾. 基于神经网络的飞机发动机性能趋势预测系统[ J]. 中国制造业信息化, 2008, 37( 11): 58- 63.
[ 7] 陈效华, 魏少华, 常思勤, 等. 基于声强识别的发动机智能化故障诊断系统[ J]. 南京理工大学学报(自然科学版), 2005, 29( 1): 36- 39.
[ 8] Co lorni A, DorigoM, M an iezzo V, e t a .l Distributedoptim ization by ant co lon ies [ A ]. Proceed ing s o f theFirst European Conference on A rtific ia l Life[ C ]. Pa ris:E lsev ie r Pub lishing, 1991: 134- 142.
[ 9] 陈恬, 孙健国, 杨蔚华, 等. 自组织神经网络航空发动机气路故障诊断[ J]. 航空学报, 2003, 24( 1): 46- 48.
[ 10] 陈果, 宋兰琪, 陈立波. 基于神经网络规则提取的航空发动机磨损故障诊断知识获取[ J]. 航空动力学报, 2008, 23( 12): 2170- 2176.
[ 11] 王志, 艾延廷, 沙云东. 基于BP神经网络的航空发动机整机振动故障诊断技术研究[ J]. 仪器仪表学报, 2007, 28( 4) : 168- 171.
[ 12] 陈恬, 孙健国, 郝英. 基于神经网络和证据融合理论的航空发动机气路故障诊断[ J]. 航空学报,2006, 27( 6): 46- 48.

Memo

Memo:
-
Last Update: 2012-11-02