基于机器学习的间歇式电沉积制备W@Cu核-壳粉体模型的构建与应用

Construction and application of model of W@Cu core-shell powder prepared by intermittent electrodeposition technology based on machine learning

W和Cu两相的均匀分布对获得高性能W-Cu复合材料至关重要。本文主要研究基于机器学习的间歇式电沉积制备W、Cu均匀分布的W@Cu粉体模型的构建与应用。首先,建立间歇式电沉积制备W@Cu核-壳粉体的机器学习模型,确定核-壳粉体理论镀层厚度与电流、电沉积时间、待镀粉体粒径和承载量之间的关联,然后在承载量为1000 g的装置中进行实验验证。将W@Cu核-壳粉体在1375℃下进行无压烧结,研究成形压力对W-Cu复合材料致密度、烧结收缩率和电导率的影响。结果表明,在电流密度为7 A/dm^(2)、电沉积时间为6 h时,实际镀层厚度为3.93μm,与理论镀层厚度3.15μm相符。提高成形压力有利于获得高致密度、低烧结收缩率的W-Cu复合材料。同时,核-壳粉体在显微组织中形成Cu的导电通道,有利于复合材料电导率的提升。

The uniform distribution of W and Cu phases is very important for obtaining high performance W-Cu composites.This paper mainly studies the construction and application of the model of W@Cu core-shell powder prepared by intermittent electrodeposition based on machine learning.Firstly,a machine learning model for the preparation of W@Cu core-shell powder by intermittent electrodeposition was established,and the relationships between the theoretical core-shell powder coating thickness and current,electrodeposition time,particle size of the powder to be plated,and the load capacity were determined.Further,the experimental verification was carried out in a device with a capacity of 1000 g.The W@Cu core-shell powders were sintered at 1375℃without pressure,and the effects of forming pressure on the density,sintering shrinkage,and electrical conductivity of W-Cu composite were investigated.The results show that the actual coating thickness is 3.93μm when the current density is 7 A/dm^(2)and the deposition time is 6 h,which is consistent with the theoretical coating thickness(3.15μm).Increasing the forming pressure is beneficial to obtain the compact W-Cu composites with low sintering shrinkage.At the same time,the core-shell powder formed a Cu conduction channel in the microstructure,which is conducive to improve the conductivity of the composite.