用BP神经网络和遗传算法优化电沉积Cu-W的工艺参数  

Process Parameters Optimization of Cu-W Electrodeposition Based on BP Neural Network and Genetic Algorithm

作　　者：李远会[1,2] 郭忠诚[2] 万明攀[1] 黄碧芳[1] 张晓燕[1] 

机构地区：[1]贵州大学材料与冶金学院,贵州贵阳550003 [2]昆明理工大学冶金与能源工程学院,云南昆明650039

出　　处：《电镀与环保》2015年第2期12-13,共2页Electroplating & Pollution Control

基　　金：国家自然科学基金(No.50964008);贵州省科学技术基金黔科合J字[2012]2114号

摘　　要：用神经网络和遗传算法(BP-GA)优化电沉积Cu-W的工艺参数。结果显示:BP-GA预测结果与试验结果较接近,相对误差为9.05%,说明BP-GA优化电沉积工艺参数有较高的预测能力和准确度。The process parameters of Cu-W electrodeposition were optimized based on BP neural network and genetic algorithm. Results showed that prediction result was approach to experimental result,and the relative error was 9.05%,which indicating that BP neural network and genetic algorithm have high forecasting accuracy and was suitable for electrodeposition parameters prediction.

关 键 词：BP-GA 电沉积 Cu-W 工艺参数 

分 类 号：TQ153[化学工程—电化学工业]