摘要
采用复合电镀工艺制备了Cu-Al_2O_3复合镀层。用X射线衍射仪表征了镀层的微观结构,计算出平均晶粒尺寸,并用维氏硬度计测量了镀层的显微硬度。将电流密度和镀液中Al_2O_3微粒的质量浓度作为输入变量,并将镀层的平均晶粒尺寸和显微硬度作为输出结果,建立了RBF神经网络。仿真结果表明:RBF神经网络具有较强的预测能力,其预测结果与实测结果较为接近,平均误差约为0.15%,为Cu-Al_2O_3复合电镀工艺优化提供了参考。
Cu-Al_2O_3 composite coatings were prepared by composite electroplating technology.The microstructure of the coatings was characterized by X-ray diffractometer,and the average grain size was calculated,the microhardness of the coatings was also measured by vickers hardness tester.RBF neural network was established by using current density and mass concentration of Al_2O_3 particles in plating bath as the input variable,and the average grain size and microhardness of the coatings as output results.The simulation results showed that RBF neural network has preferable predictive ability,the predicted value and experimental value were in good agreement,and the average error was about 0.15%.It provides references for the optimization of Cu-Al_2O_3 composite electroplating technology.
引文
[1]李家明,徐淑庆,梁铭忠.复合电镀机制的研究进展[J].电镀与环保,2016,36(2):1-3.
[2]杨国为,王守觉,闫庆旭.分式线性神经网络及其非线性逼近能力研究[J].计算机学报,2007,30(2):189-199.
[3]邱占芝,李世峰.基于神经网络的PID网络化控制系统建模与仿真[J].系统仿真学报,2018,30(4):1423-1432.
[4]黄碧芳,付维松,王丽.基于神经网络优化Cu-W-Ni电镀工艺[J].现代机械,2014(3):31-33.
[5]潘迪.神经网络模式识别及其实现[M].北京:电子工业出版社,1999.
[6]朴楠,陈吉,孙彦伟,等.脉冲电流密度对纳米晶镍镀层结构及性能的影响[J].电镀与环保,2016,36(2):4-7.
[7]刘鹏,陈登福,何文杰.Ni/Al2O3复合镀层中Al2O3微粒尺寸及热处理对镀层性能的影响[J].表面技术,2017,46(11):230-236.