小样本下双相钢物理基晶体塑性本构模型参数的多阶段渐进式识别方法研究

doi:10.3901/JME.260561

摘要/Abstract

摘要： 基于物理机制的晶体塑性本构模型被广泛应用于研究双相钢的塑性行为，但由于双相钢组元相的机械性能难以单独确定，且物理基晶体塑性本构模型参数间存在强耦合性，导致参数标定困难。为此，基于深度学习算法，针对描述双相钢塑性行为的物理基晶体塑性本构模型，构建通过材料宏观真应力-真应变曲线和不同变形阶段下的两相位错密度来识别其参数的神经网络模型。面向无织构强度的双相钢，结合晶体塑性有限元和代表体元法建立单向拉伸数值模拟模型，通过仿真计算获得模型训练所需样本库。为解决常规训练所得神经网络系统对本构参数的识别效果差问题，提出多阶段渐进式识别法；首先对参数进行解耦，而后，根据参数对材料性能影响的大小，分阶段对其进行识别，在提高识别效果的同时，有效减少所需样本量。验证集数据和试验数据测试结果表明，所提出的多阶段渐进式参数识别方法可在小样本条件下准确获得双相钢物理基晶体塑性本构模型参数。该方法为双相钢组元相晶体塑性本构模型参数的准确识别提供了可行的路径。

关键词: 晶体塑性, 双相钢, 参数识别, 小样本, 多阶段渐进式识别

Abstract: The physically-based crystal plasticity constitutive model is widely applied to study the plastic behavior of duplex steels. However，due to the difficulty in separately determining the mechanical properties of the constituent phases of duplex steels，and the strong coupling between the parameters of the physically-based crystal plasticity model，parameter calibration becomes challenging. To address this issue，based on deep learning algorithms，a neural network model is constructed to identify the parameters of a physically-based crystal plasticity model describing the plastic behavior of duplex steels. The model utilizes the macroscopic true stress-true strain curve of the material and the dislocation densities of the two phases at different deformation stages as input data. For duplex steels without texture strengthening，a uniaxial tensile numerical simulation model is developed by combining crystal plasticity finite element analysis and the representative volume element method. The simulation results provide the sample database required for model training. To overcome the poor parameter identification performance typically observed in conventional neural networks，a multi-stage progressive identification method is proposed. The method first decouples the parameters，then identifies them in stages based on their impact on material properties. This approach improves the identification accuracy while effectively reducing the required sample size. Validation test results，including both validation set data and experimental data，demonstrate that the proposed multi-stage progressive parameter identification method can accurately obtain the physically-based crystal plasticity model parameters for duplex steel under small sample conditions. This method provides a feasible approach for the accurate identification of the parameters of the crystal plasticity constitutive model in duplex steels.

Key words: crystal plasticity, duplex steel, parameter identification, small sample, multi-stage progressive identification

中图分类号:

TG156

李小龙, 吴昊芃, 秦锐, 陈雷, 金淼, 张庆玲. 小样本下双相钢物理基晶体塑性本构模型参数的多阶段渐进式识别方法研究[J]. 机械工程学报, 2026, 62(12): 163-172.

LI Xiaolong, WU Haopeng, QIN Rui, CHEN Lei, JIN Miao, ZHANG Qingling. Research on a Multi-stage Progressive Identification Method for the Constitutive Parameters of the Duplex Steels′s Physical-based Crystal Plasticity Model under Small Sample Conditions[J]. Journal of Mechanical Engineering, 2026, 62(12): 163-172.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260561

http://www.cjmenet.com.cn/CN/Y2026/V62/I12/163

参考文献

[1] ANSSI L，SAMPO U，MATTI L，et al. Micromechanics driven design of ferritic-austenitic duplex stainless steel microstructures for improved cleavage fracture toughness[J]. Engineering Fracture Mechanics，2021，253:107878.
[2] LEE M G，KIM S J，HAN H N. Crystal plasticity finite element modeling of mechanically induced martensitic transformation (MIMT) in metastable austenite[J]. International Journal of Plasticity，2010，26:688-710.
[3] LU Damin，ZHANG Keshi，HU Guijuan. Investigation on plastic flow behaviors of FCC polycrystalline aluminum under pre-cyclic tension-compression loading:Experiments and crystal plasticity modeling[J]. Nanomaterials，2021，11:2379.
[4] AHNA D H，LEEA G G，MOON J，et al. Analysis of texture and grain shape effects on the yield anisotropy of Zr-2.5wt% Nb pressure tube alloy using crystal plasticity finite element method[J]. Journal of Nuclear Materials，2021，555:153112.
[5] ZHANG Wenbin，JIN Miao，HAO Shuo，et al. The interfacial damage of the deformation heterogeneity in the transformation-induced plasticity (TRIP)-assisted duplex stainless steel[J]. International Journal of Plasticity，2025，184:104209.
[6] SEDIGHIANI K，TRAKA K，ROTERS F，et al. Determination and analysis of the constitutive parameters of temperature-dependent dislocation-density-based crystal plasticity models[J]. Mechanics of Materials，2022，164:104117.
[7] MA A，ROTERS F. A constitutive model for fcc single crystals based on dislocation densities and its application to uniaxial compression of aluminium single crystals[J]. Acta Materialia，2004，52:3603-3612.
[8] LIU Wenqi，LIAN Junhe，ARAVAS N，et al. A strategy for synthetic microstructure generation and crystal plasticity parameter calibration of fine-grain-structured dual-phase steel[J]. International Journal of Plasticity，2020，126:102614.
[9] MOALLEMI M，KIM S J，ABBAS Z H，et al. Strain hardening analysis and deformation micromechanisms in high strength-high ductility metastable duplex stainless steels:Role of sustained stacking faults in the work hardening[J]. Materials Characterization，2023，197:112662.
[10] 郑学斌，韩龙帅，李学涛，等. DP780双相钢U弯回弹预测影响因素研究[J]. 机械工程学报，2024，60(6):198-206. ZHENG Xuebin，HAN Longshuai，LI Xuetao，et al. Study on influencing factors of u-bending springback prediction of DP780 dual-phase steel[J]. Journal of Mechanical Engineering，2024，60(6):198-206.
[11] SAIDI P，PIRGAZIC H，SANJARI M，et al. Deep learning and crystal plasticity:A preconditioning approach for accurate orientation evolution prediction[J]. Computer Methods in Applied Mechanics and Engineering，2022，389:114392.
[12] ZHANG Wenbin，HAO Shuo，LI Xiaolong，et al. The effect of strain-induced martensitic transformation (SIMT) on the heterogeneous deformation behavior among constituent phases in TRIP-assisted duplex stainless steel[J]. Journal of Materials Processing Technology，2023，318:118039.
[13] 吴洋. 基于神经网络的AV本构模型参数确定方法研究[D]. 秦皇岛:燕山大学，2022. WU Yang. Study on parameter determination method of AV constitutive model based on neural network[D]. Qinghuangdao:Yanshan University，2022.
[14] ZU R L，ZHAO J Y，LIU Z W，et al. Study on high throughput inversion method for anisotropic material parameters based on nanoindentation[J]. Experimental Mechanics，2023，63:1157-1170.
[15] MIYAZAWA Y，BRIFFOD F，SHIRAIWA T，et al. Prediction of cyclic stress-strain property of steels by crystal plasticity simulations and machine learning[J]. Materials，2019，12:3668.
[16] 刘军，王兆明，刘昌孝. 神经元输入输出关系及网络计算特性研究[J]. 电子科技大学学报，1992，21(2):127-132. LIU Jun，WANG Zhaoming，LIU Changxiao. A study of neuron input-output relationships and network computational characterization[J]. Journal of University of Electronic Science and Technology，1992，21(2):127-132.
[17] CHENG Jiahao，GHOSH Somnath. A crystal plasticity FE model for deformation with twin nucleation in magnesium alloys[J]. International Journal of Plasticity，2015，67:148-170.
[18] 章海明. 位错密度晶体塑性模型及在微成形模拟中的应用[D]. 上海:上海交通大学，2013. ZHANG Haiming. Dislocation density based crystal plasticity finite element method for the simulation of micro-forming processes[D]. Shanghai:Shanghai Jiao Tong University，2013.
[19] AKHTAR S，LIU Jian，YOON J W，et al. Strain rate effect of high purity aluminum single crystals:Experiments and simulations[J]. International Journal of Plasticity，2015，67:39-52.
[20] SEDIGHIANI K，DIEHL M，TRAKA K，et al. An efficient and robust approach to determine material parameters of crystal plasticity constitutive laws from macro-scale stress-strain curves[J]. International Journal of Plasticity，2020，134:102779.
[21] 聂君锋，汤镇睿，张海泉，等. 基于位错密度的体心立方晶体塑性本构模型[J]. 清华大学学报，2017，57(7):780-784. NIE Junfeng，TANG Zhenrui，ZHANG Haiquan，et al. Crystal plasticity constitutive model for BCC based on the dislocation density[J]. J Tsinghua University，2017，57(7):780-784.
[22] ZHANG Wenbin，GUO Shu，LIU Shengjie，et al. Quantitatively assessing the contributions of temperature-dependent deformation-induced martensitic transformation to uniform elongation and work hardening of TRIP-assisted duplex stainless steel via crystal plasticity[J]. Materials Science & Engineering A，2023，887:145758.
[23] JIANG Maoyuan，FAN Zhengxuan，KRUCH S，et al. Grain size effect of FCC polycrystal:A new CPFEM approach based on surface geometrically necessary dislocations[J]. International Journal of Plasticity，2022，150:103181.
[24] LEE M G，LIM H，ADAMS B L，et al. A dislocation density-based single crystal constitutive equation[J]. International Journal of Plasticity，2010，26:925-938.
[25] 蔡旺，王春晖，孙朝阳，等. 耦合温度效应晶体塑性的TWIP钢变形行为研究[J]. 机械工程学报，2022，58(20):283-293. CAI Wang，WANG Chunhui，SUN Chaoyang，et al. Investigation of deformation behavior of TWIP steel using crystal plasticity coupling temperature effect[J]. Journal of Mechanical Engineering，2022，58(20):283-293.
[26] ZHANG Haiming，DONG Xianghuai，DU Dapeng，et al. A unified physically based crystal plasticity model for FCC metals over a wide range of temperatures and strain rates[J]. Materials Science & Engineering A，2013，564:431-441.
[27] GUI Yang，AN Dayong，HAN Fengbo，et al. Multiple-mechanism and microstructure-based crystal plasticity modeling for cyclic shear deformation of TRIP steel[J]. International Journal of Mechanical Sciences，2022，222:107269.
[28] PITTS S A，JIANG Wen，DAVIDE Pizzocri，et al. A continuum dislocation dynamics crystal plasticity approach to irradiated body centered cubic α-Iron[J]. Journal of Engineering Materials and Technology，2022，144:011018.
[29] 何宽. 基于BP神经网络的晶体塑性本构参数预测方法研究[D]. 秦皇岛:燕山大学，2023. HE Kuan. Study on prediction method of crystal plastic constitutive parameters based on BP neural network algorithm[D]. Qinghuangdao:Yanshan University，2023.
[30] 章海明，徐帅，李倩，等. 晶体塑性理论及模拟研究进展[J]. 塑性工程学报，2020，27(5):12-32. ZHANG Haiming，XU Shuai，LI Qian，et al. Progress of crystal plasticity theory and simulations[J]. Journal of Plasticity Engineering，2020，27(5):12-32.
[31] TOMOTA Y，LUKAS P，HARJO S，et al. In situ neutron diffraction study of IF and ultra low carbon steels upon tensile deformation[J]. Acta Materialia，2003，51(3):819-830.