基于BP神经网络与遗传算法的镍-钴合金电镀工艺参数优化  

Optimization of Process Parameters for Electroplating of Nickel-Cobalt Alloy Coatings Based on BP Neural Network and Genetic Algorithm

作　　者：蔡静 CAI Jing(Engineering Training Center,Southwest Petroleum University,Chengdu 610500,China)

机构地区：[1]西南石油大学工程训练中心,四川成都610500

出　　处：《电镀与环保》2018年第6期50-53,共4页Electroplating & Pollution Control

摘　　要：建立了三层BP神经网络,并将遗传算法引入BP神经网络模型中,以L16(44)正交试验的数据作为训练样本,建立电镀工艺参数与镍-钴合金镀层显微硬度之间的映射关系。以显微硬度达到最大值为优化目标,运用BP神经网络与遗传算法对电镀工艺参数进行优化。在给定的电镀工艺参数范围内,得出显微硬度达到最大值时对应的电镀工艺参数为:电流密度2A/dm2,镀液pH值4.0,氨基磺酸钴50g/L,镀液温度50℃。A three layers of BP neural network was bulit, and genetic algorithm was introduced into BP neural network model. The L16 （4S）-orthogonal experimental data was chosen as the trained samples, and the mapping relation between electroplating process parameters and micro hardness of nickel cobalt alloy coating was built. In order to obatin the maximum micro hardness, the electroplating process parameters was optimized by BP neural network and genetic algorithm. In the selected electroplating process parameters range, it was concluded that the electroplating process parameters corresponding to maximum micro hardness were as follows： current density 2 A/dm2 , bath pH value 4. 0, mass concentration of cobalt sulfamate 50 g/L, bath temperature 50℃.

关 键 词：电镀工艺参数优化 镍钴合金镀层 BP神经网络 遗传算法 

分 类 号：TQ153[化学工程—电化学工业]