真空悬浮熔炼的机理-AI双闭环智能控制新范式及验证*

doi:10.13385/j.cnki.vacuum.2026.04.06

摘要/Abstract

摘要： 真空悬浮熔炼炉具备无接触加热、高纯净度和低污染等优势,广泛应用于高性能金属材料制备。该过程涉及多物理场耦合与工艺参数的非线性映射,关系复杂,传统依赖经验与试验的方法难以实现精准控制。本文构建基于热力学与电磁学原理的机理模型,结合历史运行数据修正了边界条件与参数,形成了面向熔炼过程的数字孪生体,实现了虚实一致的动态建模与更新。在此基础上,引入物理信息机器学习方法,在物理边界约束下训练AI工艺参数预测模型。将AI预测参数用于实际熔炼,所得数据反向优化机理模型与AI模型,形成了闭环优化框架。该框架显著提升了机理模型精度与AI预测可靠性,实现了工艺自适应控制与智能化生产,突破了传统经验控制与纯数据驱动模型的局限,构建了机理可解释与AI自适应相融合的智能控制新范式,为高端真空冶金装备的数字孪生、智能化升级及行业整体技术进步提供了参考。

关键词: 真空悬浮熔炼炉, 数字孪生, 人工智能, 工艺参数预测

Abstract: Vacuum levitation melting furnaces offer advantages such as contactless heating, high purity, and low pollution, making them widely used in the preparation of high-performance metal materials. However, the complex nonlinear relationships between multi-physics coupling and process parameters within them make precise control difficult for traditional methods that rely on experience and experimentation. This paper constructs a mechanistic model based on thermodynamics and electromagnetics. By combining historical operating data to modify boundary conditions and parameters, it created a digital twin of the melting process, achieving dynamic modeling and updating that was consistent between the virtual and the real. Furthermore, a physicalinformed machine learning approach was introduced to train an AI-powered process parameter prediction model under physical boundary constraints. The AI-predicted parameters were applied to actual melting, and the resulting data was used to inversely optimize the mechanistic and AI models, forming a closed-loop co-evolutionary frame work. This approach significantly improves the accuracy and prediction reliability of the mechanistic model, enabling adaptive process control and intelligent production. It transcends the limitations of traditional empirical control and purely data-driven models to construct a novel intelligent control paradigm integrating interpretable mechanisms and AI adaptability. This approach provides an effective solution for the digital twinning and intelligent upgrading of high-end vacuum metallurgical equipment, as well as for overall technological advancement in the industry.

Key words: vacuum levitation melting furnace, digital twin, artificial intelligence, process parameter prediction

中图分类号: TF743

刘志中, 曹大年, 吴韬, 罗森, 王金让, 何永亮, 张鹏程, 郝学卓, 李万青. 真空悬浮熔炼的机理-AI双闭环智能控制新范式及验证^*[J]. 真空, 2026, 63(4): 39-45.

LIU Zhizhong, CAO Danian, WU Tao, LUO Sen, WANG Jinrang, HE Yongliang, ZHANG Pengcheng, HAO Xuezhuo, LI Wanqing. Mechanism of Vacuum Suspension Melting-A New Paradigm of AI Dual Closed-Loop Intelligent Control and Its Verification[J]. VACUUM, 2026, 63(4): 39-45.

参考文献

[1] Boikov A, Payor V.The present issues of control automation for levitation metal melting[J]. Symmetry, 2022, 14(10):1968.
[2] Golak S, Wyciślik J, Zybała R, et al.Numerical modeling and optimization of large-scale molten titanium levitation[J]. Materials, 2025, 18(6):1268.
[3] Nycz B, Przyłucki R, Maliński Ł, et al.Optimization of an inductor for electromagnetic levitation melting[J]. COMPEL-The international journal for computation and mathematics in electrical and electronic engineering, 2024, 43(6):1103-1112.
[4] Nycz B, Malinski L, Przylucki R.Influence of selected model parameters on the electromagnetic levitation melting efficiency[J]. Applied Sciences, 2021, 11(9):3827.
[5] Smalcerz A, Blacha L, Barglik J, et al.Electromagnetic processing of molten copper alloys in the induction furnace with cold crucible[J]. Archives of Electrical Engineering, 2024, 73(1):51-62.
[6] Zhang Chaojun, Cao Fuyang, Zhang Lunyong, et al.Break the superheat temperature limitation of induction skull melting technology[J]. Applied Thermal Engineering, 2023, 220:119780.
[7] Zhang Chaojun, Zhang Lunyong, Cao Fuyang, et al.Optimizing energy efficiency in induction skull melting process:investigating the crucial impact of melting system structure[J]. Scientific Reports, 2024, 14(1):6303.
[8] Wang Gangquan, Bai Yinxue, Zhao Baolin, et al.Improve the performance of cold crucible used for emissivity measurement of molten metal[J]. Case Studies in Thermal Engineering, 2025, 68:105898.
[9] Zorzetto M, Torchio R, Lucchini F, et al.Proper orthogonal decomposition for parameterized macromodeling of a longitudinal electromagnetic levitator[J]. IEEE Transactions on Magnetics, 2025,61(4):8300207.
[10] Karniadakis G E, Kevrekidis I G, Lu L, et al.Physics-informed machine learning[J]. Nature Reviews Physics, 2021, 3(6):422-440.
[11] Sharma A, Kosasih E, Zhang Jie, et al.Digital twins:state of the art theory and practice, challenges, and open research questions[J]. Journal of Industrial Information Integration, 2022, 30:100383.
[12] Wu Yuandi, Sicard B, Gadsden S A.Physics-informed machine learning:a comprehensive review on applications in anomaly detection and condition monitoring[J]. Expert Systems with Applications, 2024, 255:124678.
[13] Mendi-Altube A, Villar I, Carretero C, et al.Electro-thermal modeling of an induction heating process of 42CrMo4 steel probe[J]. International Journal of Applied Electromagnetics and Mechanics, 2024, 75(2):169-178.
[14] Baker E B, Nawer J, Xiao X, et al.MHD surrogate model for convection in electromagnetically levitated molten metal droplets processed using the ISS-EML facility[J]. npj Microgravity, 2020, 6(1):9.
[15] Spitans S, Franz H, Baake E, et al.Large-scale levitation melting and casting of titanium alloys[J]. Magnetohydrodynamics, 2017, 53(4):633-641.
[16] Hyers R W.Modeling of and experiments on electromagnetic levitation for materials processing[D]. Massachusetts Institute of Technology, 1998.
[17] Grytsiuk V Y, Yassin M A M. Numerical modeling of coupled electromagnetic and thermal processes in the zone induction heating system for metal billets[J]. Electrical Engineering & Electromechanics,2025(2):59-68.
[18] Chen Yuchao, Luo Qingxuan, Silaen A K, et al.Multi-physics modeling of steel ingot melting by electric arc plasma and its application to electric arc furnace[J]. Frontiers in Materials, 2020,7:576831.
[19] Garcia-Michelena P, Ruiz-Reina E, Herrero-Dorca N, et al.Multiphysics modeling and experimental validation of heat and mass transfer for the vacuum induction melting process[J]. Applied Thermal Engineering,2024,243:122562.
[20] Nycz B, Przyłucki R, Maliński Ł, et al.A simulation model for the inductor of electromagnetic levitation melting and its validation[J]. Materials,2023, 16(13):4634.
[21] Garcia-Michelena P, Ruiz-Reina E, Gordo-Burgoa O, et al.Numerical simulation of free surface deformation and melt stirring in induction melting using ALE and level set methods[J]. Materials, 2025,18(1):199.
[22] 章复中,谷力军,瞿柳兴.8千瓦真空电磁悬浮熔炼炉的参数选择和结构[J].真空科学与技术,1986,(Z1):27-31.
[23] 闫学顺,汪东红,吴文云,等. 熔模铸造车间数据采集与生产智能预测[J]. 特种铸造及有色合金,2024,44(1):135-140.
[24] Giannetti C, Borghini E, Carr A, et al.Deep learning for robust forecasting of hot metal silicon content in a blast furnace[J]. The International Journal of Advanced Manufacturing Technology, 2025,136(9):4045-4054.
[25] Li Jingchang, Zhang Xiaoge, Zhou Qi, et al.A feature-level multi-sensor fusion approach for in-situ quality monitoring of selective laser melting[J]. Journal of Manufacturing Processes, 2022,84:913-926.
[26] Westad F, Lodgaard L, Pedersen T.Data Pre-processing and sensor-fusion for multivariate statistical process control of an extrusion process[C]//Proceedings of the 3rd International Workshop on Software Engineering and AI for Data Quality in Cyber-Physical Systems/Internet of Things. 2023:9-15.
[27] Derouiche K, Daoud M, Traidi K, et al.Real-time prediction by data-driven models applied to induction heating process[J]. International Journal of Material Forming, 2022,15(4):48.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed