摘要
根据现有硬度预测模型各自的主要特点,其可以分为经验模型、基于材料强化机制的预测模型、引入多变量的预测模型和采用不同预测方法的模型共4大类。前3类模型均为显式的数学模型,可直接反映不同变量与硬度间的关系,物理意义较为明确;第4类模型则更容易获得高精度的预测结果。考虑相互影响的多变量硬度预测模型具有良好的研究前景,而采用神经网络或支持向量回归方法的硬度预测模型具有良好的应用前景。
According to the main characteristic,the current prediction models of hardness can be divided into four categories: empirical models,prediction models based on material strengthening mechanism,prediction models considering multiple variables and models based on different prediction methods. The first three types of models are all directly expressed in mathematical way,in which the relationship between different variables and hardness can be reflected directly,and the physical significance is relatively clear. The fourth kind of models is easier to achieve prediction results with high accuracy. The hardness prediction models considering the interaction of multiple variables have good research prospect,and the hardness prediction models based on neural network or support vector regression have good application prospect.
作者
胡成亮
胡誉
张运军
邓庆文
权开峰
李生仕
赵震
HU Cheng-liang;HU Yu;ZHANG Yun-jun;DENG Qing-wen;QUAN Kai-feng;LI Sheng-shi;ZHAO Zhen(National Engineering Research Center of Die&Mold CAD,Shanghai Jiao Tong University,Shanghai 200030,China;Institute of Forming Technology and Equipment,Shanghai Jiao Tong University,Shanghai 200030,China;Hubei Tri-ring Forging Co.,Ltd.,Xiangyang 441700,China)
出处
《塑性工程学报》
CAS
CSCD
北大核心
2020年第5期7-11,共5页
Journal of Plasticity Engineering
基金
工业强基工程项目(TC180A3Y1/18)
国家自然科学基金资助项目(51875348)。
关键词
硬度
预测模型
晶粒尺寸
位错密度
等效应变
hardness
prediction model
grain size
dislocation density
effective strain
