摘要
金属材料在冲击、爆炸等高应变率加载下的塑性流动行为具有不同于静载下的率-温耦合性和微观机制。航空航天、航海、能源开采、核工业、公共安全、灾害防治等方面的金属结构设计与性能评估需要进行大量的动载实验和数值模拟,建立准确的材料动态本构模型是结构数值模拟可靠性的基础和关键。本文中,总结了金属材料的率-温耦合变形行为及内在机理,回顾了金属动态本构关系研究的起源与发展脉络,分别针对唯象模型、具有物理基础的模型和人工神经网络模型进行了详细介绍和横向比较。唯象模型和人工神经网络模型分别因易应用和高预测精度而受到青睐,基于物理概念的宏观连续介质模型可以描述体现内部演化的真实物理量,从而涵盖更大的应变范围,更好地反映应变率、温度和应变的影响机制。
Different from static loading conditions, the plastic flow behavior of metallic materials under high strain rate loadings, such as impact and explosion, exhibits special rate-temperature coupling effect and deformation micro-mechanism.The design and evaluation of metallic structures used in aerospace and navigation, energy mining, nuclear industry, public safety, disaster prevention, etc. require a large number of experiments under dynamic loadings. In recent years, the rapiddeveloping computational mechanics can be used to analyze the structural mechanical response under complex loading,evaluate the structural safety and optimize the structural design, and can also save the experimental costs. Accurate dynamic constitutive description of materials is the basis for the reliability of structural numerical simulation. In this paper, the dynamic plastic deformation behavior and micro-mechanism of metals, as well as the origin and development of the dynamic constitutive relationship of metals are reviewed and summarized. Over wide ranges of strain rate and temperature, the metals exhibit complex rate-temperature coupling effect, such as dynamic strain aging and segmented strain rate sensitivity. The high strain rate may lead to dynamic recrystallization, deformation twinning and shock-induced phase transition. The existing constitutive models can be divided into three types: phenomenological models, physically based models and artificial neural network models. Phenomenological models refer to the constitutive models established merely by describing experimental phenomena without considering the internal physical mechanism. Physically based macro-scale continuum models can represent true physical quantities for documenting and tracking the evolution which takes place within metallic materials.Artificial neural network models are good at reproducing the plastic flow behavior as function of many factors, such as strain rate, temperature and plastic strain, without the need of identifying complex logic relationships and parameters within the system. The developments, prediction capabilities, and application scopes of the three types of dynamic constitutive models are illustrated in detail and compared horizontally. In addition, some objective suggestions for the further development of dynamic constitutive descriptions for metals are proposed. Phenomenological models are favored for their ease in application, artificial neural network models are favored for their high prediction accuracy. Recent trend has increased the focus on physically based models. This type of model extends application to a wider strain range and more clearly represents the influence mechanism of strain rate, temperature and strain.
作者
袁康博
姚小虎
王瑞丰
莫泳晖
YUAN Kangbo;YAO Xiaohu;WANG Ruifeng;MO Yonghui(Department of Engineering Mechanics,South China University of Technology,Guangzhou 510641,Guangdong,China;School of Aeronautics,Northwestern Polytechnical University,Xi’an 710072,Shaanxi,China)
出处
《爆炸与冲击》
EI
CAS
CSCD
北大核心
2022年第9期2-35,共34页
Explosion and Shock Waves
基金
中央高校基本科研业务费专项资金(x2tjD2220850)
国家自然科学基金(12202149)
国家杰出青年科学基金(11925203)
中国博士后科学基金(2022M711198)。
关键词
金属材料
高应变率
塑性变形
动态本构模型
metallic material
high strain rate
plastic deformation
dynamic constitutive model
