基于正交试验设计与人工神经网络优化镀铬工艺

腐蚀与防护, 2014, 35(1): 78-82.

基于正交试验设计与人工神经网络优化镀铬工艺

钟庆阳 ^1, , 李振华 ^2, , 周琼宇 ^3, , 李珂 ^4, , 钟庆东 ^5,

1.中国石化集团江汉石油管理局勘察设计研究院,武汉430073

2.上海大学上海市现代冶金与材料制备重点实验室,上海200072

3.上海大学上海市现代冶金与材料制备重点实验室,上海200072

4.上海大学上海市现代冶金与材料制备重点实验室,上海200072

5.上海大学上海市现代冶金与材料制备重点实验室,上海200072

基金项目: 教育部创新团队计划资助项目(IRT0739) 国家自然科学基金(50571059,50615024) 汽车用钢开发与应用技术国家重点实验室(宝钢)开放课题

关键词：镀铬 , 正交试验 , 人工神经网络 , 权重值 , 工艺优化

Optimization of Chrome-plating Craft Based on Orthogonal Test Design and Artificial Neural Network

ZHONG Qing-yang ^1, , LI Zhen-hua ^2, , ZHOU Qiong-yu ^3, , LI Ke ^4, , ZHONG Qing-dong ^5,

1.中国石化集团江汉石油管理局勘察设计研究院,武汉430073

2.上海大学上海市现代冶金与材料制备重点实验室,上海200072

3.上海大学上海市现代冶金与材料制备重点实验室,上海200072

4.上海大学上海市现代冶金与材料制备重点实验室,上海200072

5.上海大学上海市现代冶金与材料制备重点实验室,上海200072

Keywords： chromium plating , orthogonal test , artificial neural network , weight value , craft optimization

提出了一种正交试验设计与人工神经网络相结合的镀铬工艺参数优化方法.样本极差结果表明,对镀铬层的厚度及阴极电流效率影响因素依次为电流密度、电镀时间、电镀温度;且最佳电镀温度为45℃.通过神经网络建立电镀工艺参数与性能之间的模型,预测得出的镀铬层的厚度和阴极电流效率与实际试验的结果接近,训练精度较高,预测值与试验值的相对误差小于1.20％.通过建立镀铬层多指标综合评价模型,对镀铬层的厚度及阴极电流效率两个指标进行综合评价,通过对两个指标权重值的调整,确定镀铬层的综合性能值,得出最优的工艺参数.

参考文献

[1]	Yu HH;Xing RG;Liu S;Li CP;Guo ZY;Li PC .Studies on the hemolytic activity of tentacle extracts of jellyfish Rhopilema esculentum Kishinouye: Application of orthogonal test[J].International Journal of Biological Macromolecules: Structure, Function and Interactions,2007(3):276-280.
[2]	Diez C;Barrado E;Marinero P;Sanz A .Orthogonal array optimization of a multiresidue method for cereal herbicides in soils[J].Journal of chromatography, A: Including electrophoresis and other separation methods,2008(1/2):10-23.
[3]	Ugo Galvanetto;George Violaris .Numerical investigation of a new damage detection method based on proper orthogonal decomposition[J].Mechanical Systems & Signal Processing,2007(3):1346-1361.
[4]	Benoit Igne;Jean-Michel Roger;Sylvie Roussel;Veronique Bellon-Maurel;Charles R. Hurburgh .Improving the transfer of near infrared prediction models by orthogonal methods[J].Chemometrics and Intelligent Laboratory Systems,2009(1):57-65.
[5]	F. Sanchez Lasheras;J. A. Vilan Vilan;P. J. Garcia Nieto;J. J. del Coz Diaz .The use of design of experiments to improve a neural network model in order to predict the thickness of the chromium layer in a hard chromium plating process[J].Mathematical and computer modelling,2010(7/8):1169-1176.
[6]	Edwina.hernandez-Caraballo;Rita M.Avila-Gomez;Francklin Rivas;marcela Burguera;Jose l.Burguera .Increasing the working calibration range by means of artificial neural networks for the determination of cadmium by graphite furnace atomic absorption spectrometry[J].Talanta: The International Journal of Pure and Applied Analytical Chemistry,2004(2):425-431.
[7]	高礼让,吴秀红,高志明,李复志,张爱莲.人工神经网络分光光度法同时测定钼和铬[J].光谱学与光谱分析,1999(02):244-246.
[8]	Rosa, E.S.;Salgado, R.M.;Ohishi, T.;Mastelari, N. .Performance comparison of artificial neural networks and expert systems applied to flow pattern identification in vertical ascendant gas-liquid flows[J].International Journal of Multiphase Flow,2010(9):738-754.

上一张下一张

计量

下载量()
访问量()

作者相关

文章评分

您的评分：
1

0%
2

0%
3

0%
4

0%
5

0%

材料期刊网