基于正交试验设计与人工神经网络优化镀铬工艺
Optimization of Chrome-plating Craft Based on Orthogonal Test Design and Artificial Neural Network
摘 要
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。
正交试验
人工神经网络
权重值
工艺优化
chromium plating
orthogonal test
artificial neural network
weight value
craft optimization
Abstract
A method combining orthogonal experimental design with artificial neural networks was proposed to optimize the chrome-plating craft parameter. The results show that the chrome-plating thickness and cathodic efficiency are influenced by the current density, galvanization time and galvanization temperature in turn; the best galvanization temperature was 45 ℃. The model between the galvanization technological parameters and the performance by artificial neural networks was established and the chrome-plating thickness and cathodic efficiency predicted by the model were closed to actual experimental results. The training precision was accurate, and the relative error between the predicted value and the experimental value was less than 1.2%. A comprehensive evaluation model was established to evaluate two indicators which were chrome-plating thickness and cathodic current efficiency. The model adjusted the weight value of two indicators respectively, calculated the chromium plating comprehensive performance value and got the optimal craft parameters.
中图分类号 TG174
所属栏目 应用技术
基金项目 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; 教育部创新团队计划资助项目(IRT0739)
收稿日期 2013/2/28
修改稿日期
网络出版日期
作者单位点击查看
备注钟庆东(1969-),教授,博士,从事材料腐蚀电化学研究,
引用该论文: ZHONG Qing-yang,LI Zhen-hua,ZHOU Qiong-yu,LI Ke,ZHONG Qing-dong. Optimization of Chrome-plating Craft Based on Orthogonal Test Design and Artificial Neural Network[J]. Corrosion & Protection, 2014, 35(1): 78
共有人对该论文发表了看法，其中：
人认为该论文很差
人认为该论文较差
人认为该论文一般
人认为该论文较好
人认为该论文很好
参考文献
【1】YU H H,XING R,LIU S,et al. Studies on the hemolytic activity of tentacle extracts of jellyfish rhopilema esculentum kishinouye:Application of orthogonal test[J]. International Journal of Biological Macromolecules,2007,40(3):276-280.
【2】PENG J,LEE C I C,LIU L,et al. Orthogonal multiple contrast test for testing monotone on the average[J]. Statistics & Probability Letters,2006,76(2):161-168.
【3】DIEZ C,BARRADO E,MARINERO P,et al. Orthogonal array optimization of a multiresidue method for cereal herbicides in soils[J]. Journal of Chromatography A,2008,1180(1/2):10-23.
【4】UGO Galvanetto,GEORGE Violaris. Numerical investigation of a new damage detection method based on proper orthogonal decomposition[J]. Mechanical Systems and Signal Processing,2007,21(3):1346-1361.
【5】BENOIT Igne,JEAN-Michel Roger,SYLVIE Roussel,et al. Improving the transfer of near infrared prediction models by orthogonal methods[J]. Chemometrics and Intelligent Laboratory Systems,2009,99(1):57-65.
【6】CUCCHETTI A,CESCON M,GRAZI G L,et al. Preoperative prediction of hepatocellular carcinoma tumor grade and micro-vascular invasion by means of artificial neural network: A Pilot Study[J]. Liver Transplantation,2010,16(6):77-78.
【7】LASHERAS F S,VILAN J A V,NIETO P J G,et al. 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,2010,52(7/8):1169-1176.
【8】HERNANDEZ C E A,AVILA G R M,RIVAS F,et al. Increasing the working calibration range by means of artificial neural networks for the determination of cadmium by graphite furnace atomic absorption spectrometry[J]. Talanta,2004,63(2):425-431.
【9】GAO L R,WU X H,GAO Z M,et al. Simultaneous determination of molybdenum and chromium by artificial neural networks spectrophotometry[J]. Spectroscopy and Spectralanalysis,1999,19(2):244-246.
【10】ROSA E S,SALHADO R M,OHISHI T,et al. Performance comparison of artificial neural networks and expert systems applied to flow pattern identification in vertical ascendant gas-liquid flows[J]. International Jouraal of Multiphase Flow,2010,36(9):738-754.
【11】Al-MASRI A,A B Kadir M Z A,HIZAM H,et al. Enhancement in static security assessment for a power system using an optimal artificial neural network[J]. International Review of Electrical Engineering-IREE,2010,5(3):1095-1102.
【2】PENG J,LEE C I C,LIU L,et al. Orthogonal multiple contrast test for testing monotone on the average[J]. Statistics & Probability Letters,2006,76(2):161-168.
【3】DIEZ C,BARRADO E,MARINERO P,et al. Orthogonal array optimization of a multiresidue method for cereal herbicides in soils[J]. Journal of Chromatography A,2008,1180(1/2):10-23.
【4】UGO Galvanetto,GEORGE Violaris. Numerical investigation of a new damage detection method based on proper orthogonal decomposition[J]. Mechanical Systems and Signal Processing,2007,21(3):1346-1361.
【5】BENOIT Igne,JEAN-Michel Roger,SYLVIE Roussel,et al. Improving the transfer of near infrared prediction models by orthogonal methods[J]. Chemometrics and Intelligent Laboratory Systems,2009,99(1):57-65.
【6】CUCCHETTI A,CESCON M,GRAZI G L,et al. Preoperative prediction of hepatocellular carcinoma tumor grade and micro-vascular invasion by means of artificial neural network: A Pilot Study[J]. Liver Transplantation,2010,16(6):77-78.
【7】LASHERAS F S,VILAN J A V,NIETO P J G,et al. 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,2010,52(7/8):1169-1176.
【8】HERNANDEZ C E A,AVILA G R M,RIVAS F,et al. Increasing the working calibration range by means of artificial neural networks for the determination of cadmium by graphite furnace atomic absorption spectrometry[J]. Talanta,2004,63(2):425-431.
【9】GAO L R,WU X H,GAO Z M,et al. Simultaneous determination of molybdenum and chromium by artificial neural networks spectrophotometry[J]. Spectroscopy and Spectralanalysis,1999,19(2):244-246.
【10】ROSA E S,SALHADO R M,OHISHI T,et al. Performance comparison of artificial neural networks and expert systems applied to flow pattern identification in vertical ascendant gas-liquid flows[J]. International Jouraal of Multiphase Flow,2010,36(9):738-754.
【11】Al-MASRI A,A B Kadir M Z A,HIZAM H,et al. Enhancement in static security assessment for a power system using an optimal artificial neural network[J]. International Review of Electrical Engineering-IREE,2010,5(3):1095-1102.
相关信息
