基于随机森林模型的复合钎料石墨烯晶体的完整性

Integrityof graphene crystals in composite solder materials based on random forest model

石墨烯增强Sn基复合钎料可改善钎料基体的润湿性、增强钎料的导热、导电性及综合力学性能。复合钎料中石墨烯的晶体结构完整性受到制备过程中粉末冶金工艺的严重影响,削弱了石墨烯的强化效果。因此,采用随机森林(RF)算法对高能球磨工艺参数与混合粉末中石墨烯晶体结构完整性的关系进行了研究,结果显示RF模型在训练集和测试集的相关系数R^(2)分别达到0.97和0.94,同时RF模型预测的球磨后的石墨烯拉曼光谱特征峰值变化趋势与透射电镜观察到的微观结构变化一致。表明RF模型在对球磨工艺与石墨烯晶体结构完整性的映射关系上有较好的预测性能。采用该模型对于石墨烯增强Sn基复合钎料的制备具有重要的指导意义。

Graphene reinforced Sn based composite brazing materials can improve the wettability of the brazing matrix,enhance the thermal conductivity,conductivity,and comprehensive mechanical properties of the brazing material.The crystal structure integrity of graphene in composite brazing materials is severely affected by the powder metallurgy process during preparation,which weakens the strengthening effect of graphene.Therefore,this article uses the Random Forest(RF)algorithm to study the relationship between high-energy ball milling process parameters and the integrity of graphene crystal structure in mixed powders.The results show that the correlation coefficients R^(2) of the RF model in the training and testing sets reach 0.97and 0.94,respectively.At the same time,the peak change trend of graphene Raman spectroscopy characteristics predicted by the RF model after ball milling is consistent with the microstructural changes observed by transmission electron microscopy.This indicates that the RF model has good predictive performance in the mapping relationship between ball milling process and the integrity of graphene crystal structure.The adoption of this model has important guiding significance for the preparation of graphene reinforced Sn based composite brazing materials.