Toward helicopter gearbox diagnostics from a small number of examples

被引:32
作者
Zacksenhouse, M [1 ]
Braun, S [1 ]
Feldman, M [1 ]
Sidahmed, M [1 ]
机构
[1] Technion Israel Inst Technol, Fac Mech Engn, IL-32000 Haifa, Israel
关键词
Accelerometers - Database systems - Failure analysis - Gears - Mathematical models - Neural networks - Shafts (machine components) - Vehicle transmissions - Vibrations (mechanical);
D O I
10.1006/mssp.2000.1297
中图分类号
TH [机械、仪表工业];
学科分类号
0802 ;
摘要
The United States Navy funded Westland Helicopters LTD to carry out a series of tests on the CH-46E helicopter transmission for the purpose of generating a database that can be used to evaluate diagnostic tools in general, and neural networks in particular. The database is extensive in the number of accelerometers, load conditions and diversity of faults. However, only 1-4 instances (levels) of each fault were included. Thus, the number of examples that can be extracted to train any classification scheme is limited, and well below the minimum required. We argue that instead the database is best used to develop specific signal generation models for the CH-46E helicopter gearbox. Such models can then be used to suggest meaningful features and derive appropriate signal-processing tools. Specifically, we demonstrate that the meshing vibrations induced by the large collector gear located on the quill shaft are significant and may interact with the vibrations induced by other elements attached to the same shaft. An appropriate model is developed and the effect, termed cross-gear-pair interaction, is studied using different signal-processing tools, (C) 2000 Academic Press.
引用
收藏
页码:523 / 543
页数:21
相关论文
共 18 条
[1]   The Vapnik-Chervonenkis Dimension: Information versus Complexity in Learning [J].
Abu-Mostafa, Yaser S. .
NEURAL COMPUTATION, 1989, 1 (03) :312-317
[2]  
[Anonymous], TIME SERIES DATA ANA
[3]   What Size Net Gives Valid Generalization? [J].
Baum, Eric B. ;
Haussler, David .
NEURAL COMPUTATION, 1989, 1 (01) :151-160
[4]  
Bishop C. M., 1995, NEURAL NETWORKS PATT
[5]   LEARNABILITY AND THE VAPNIK-CHERVONENKIS DIMENSION [J].
BLUMER, A ;
EHRENFEUCHT, A ;
HAUSSLER, D ;
WARMUTH, MK .
JOURNAL OF THE ACM, 1989, 36 (04) :929-965
[6]  
Braun S., 1986, Mechanical Signature Analysis: Theory and Applications
[7]  
BRAUN S, 1998, 3 INT C AC VIBR SURV
[8]  
CAMERON BG, WESTLAND HELICOPTER
[9]  
CAPDESSUS C, 1995, P 2 INT S AC VIBR SU, P391
[10]   A GENERAL LOWER BOUND ON THE NUMBER OF EXAMPLES NEEDED FOR LEARNING [J].
EHRENFEUCHT, A ;
HAUSSLER, D ;
KEARNS, M ;
VALIANT, L .
INFORMATION AND COMPUTATION, 1989, 82 (03) :247-261