|Table of Contents|

Automatic Identification for Ground Burning Crack of Titanium Alloy Parts Based on Genetic Algorithms and Neural Networks

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

Issue:
2006年06期
Page:
697-700
Research Field:
Publishing date:
2006-12-30

Info

Title:
Automatic Identification for Ground Burning Crack of Titanium Alloy Parts Based on Genetic Algorithms and Neural Networks
Author(s):
WANG Zong-rong1 2 ZUO Dun-w en1 WANG M in1
1. Mechan ical and E lectrical College, N anjingUn iversity o fAeronaut ics and A stronautics, N anjing 210016, Ch ina; 2. Department o fD ig ita lManufacturing, Yancheng Institute o fTechno logy, Yancheng 224003, Ch ina
Keywords:
titanium a lloy g round burn ing surface image processing genetic algorithms neural netw orks image recognition
PACS:
TG 506
DOI:
-
Abstract:
In order to reso lve problems ofmon itoring and forecast ing on line for g round burn ing crack of parts. materia l d ifficu lt to cut such as t itan ium a lloy, an automat ic identif ication algor ithm is pu t forw ard for ground burn crack of t itan ium alloy parts based on neura l netw orks and image-processing techno logy. In th is method, the images o f ground surface are gotten from co lour CCD. The images from CCD are transformed into gray images and into b inary images. The binary images are compressed and encoded from data po int array to pattern characterist ics. Mu lt-i layer forw ard neural ne-t w orks are tra ined by comb in ing back-propagated netw ork and genet ic algo rithms. The burning states of ground surface images are iden tified by the tra ined neura l netwo rks. The simu lat ion results show that the gained netw ork has stead iness, fast study convergence speed, strongm emory and generalizat ion ab ility. Total efficiency in this method is about 88% .

References:

[ 1] 任敬心, 康仁科, 吴小玲, 等. 钛合金的磨削烧伤和 磨削裂纹[ J]. 制造技术与机床, 2000, 28 ( 10) : 40 - 42.
[ 2] 任敬心, 康仁科, 吴小玲, 等. 钛合金的磨削裂纹及 抑制措施[ J] . 机械工人(冷加工), 2000, 51 ( 7): 5 - 6.
[ 3] 王威廉, 荆长生, 武智慧. 缓进给磨削温度、磨削烧 伤及表面层元素变化[ J]. 金刚石与磨料磨具工 程, 1987, 7( 1): 23- 25.
[ 4] Loe D, Tavaov P. H ow to gr inding titanium [ J]. Am er icanM achinist, 1954, 78( 21): 12- 13.
[ 5] Y ang C, ShawM. The g rinding of titan ium alloys[ J]. Trans o f ASME, 1955, 77( 7) : 18- 20.
[ 6] 王庆有. CCD应用技术[M ]. 天津: 天津大学出版 社, 2000. 55- 60.
[ 7] 陈国良, 王煦法, 庄镇泉, 等. 遗传算法及其应用 [M ]. 北京: 人民邮电出版社, 1996. 75- 87.
[ 8 ] 黎明, 严超华, 刘高航. 算法优化前向神经网络结 构和权重[ J] . 中国图像图形学报, 1999, 4( 6) : 462 - 466.
[ 9 ] 梁祯, 何新贵, 董德双. 前向神经网络的一种超线 性收敛BP算法[ J] . 软件学报, 2000, 11( 8): 1 094 - 1 096.

Memo

Memo:
-
Last Update: 2006-12-30