Yizheng Wang

Yizheng Wang

Yizheng Wang

Hi, my name is Yizheng Wang. I am a first year Ph.D. student in Department of Engineering Mechanics, Tsinghua University (2024-now), supervised by Prof. Yinghua Liu and Prof. Rabczuk. I major in engineering mechanics, mainly doing computational solid mechanics and received my masterโ€™s degree supervised by Prof. Yinghua Liu (2019-2022) from the Institute of Solid Mechanics (ISM), Department of Engineering Mechanics, Tsinghua University, and Bachelorโ€™s degree from Civil Aviation University of China in Tianjin. I also worked as a research assistant at Microsoft Research AI4Science (2022-2023). I have came to Bauhaus University as a guest researcher in Weimar supervised by Prof. Rabczuk for AI for mechanics (2023-2024).

I am majoring in solid mechanics and have rich experience in physical theory, computation, and experiment. I mainly focus on solving Partial Differential Equations of solid mechanics based on the physics-informed neural networks(PINNs), operator learning and deep energy method. My research interest is AI4Science, specifically AI for mechanics.

My representative works are: KINN, DCEM, CENN, and review of AI4PDEs.

๐Ÿ”ฅ News

CMAME
sym

Kolmogorov Arnold Informed neural network: A physics-informed deep learning framework for solving PDEs based on Kolmogorov Arnold Networks

Yizheng Wang, Jia Sun, Jinshuai Bai, Cosmin Anitescu, Mohammad Sadegh Eshaghi, Xiaoying Zhuang, Timon Rabczuk, and Yinghua Liu

PDF

  • We propose different PDE forms based on KAN instead of MLP, termed Kolmogorov-Arnold-Informed Neural Network (KINN). We systematically compare MLP and KAN in various numerical examples of PDEs, including multi-scale, singularity, stress concentration, nonlinear hyperelasticity, heterogeneous, and complex geometry problems. Our results demonstrate that KINN significantly outperforms MLP in terms of accuracy and convergence speed for numerous PDEs in computational solid mechanics, except for the complex geometry problem. This highlights KINNโ€™s potential for more efficient and accurate PDE solutions in AI for PDEs.
IJNME
sym

DCEM๏ผš A deep complementary energy method for solid mechanics using minimum complementary energy principle

Yizheng Wang, Jia Sun, Timon Rabczuk, and Yinghua Liu

PDF

  • We propose the deep complementary energy method (DCEM) based on the principle of minimum complementary energy for the first time. DCEM outperforms DEM in terms of stress accuracy and efficiency and has an advantage in dealing with complex displacement boundary conditions. A deep complementary energy operator method (DCEM-O) by combining operator learning with physical equations is proposed.
Advances in Mechanics
sym

AI for PDEs in solid mechanics: A review

Yizheng Wang, Xiaoying Zhuang, Timon Rabczuk, Yinghua Liu

PDF

  • We review AI for PDEs in solid mechanics.
CMAME
sym

CENN: Conservative energy method based on neural networks with subdomains for solving variational problems involving heterogeneous and complex geometries

Yizheng Wang, Jia Sun, Wei Li, Zaiyuan Lu, Yinghua Liu

PDF

  • A deep energy method with subdomains, suitable to solve non-uniform problems with complex boundaries.
IJMSD
sym

Transfer Learning in Physics-Informed Neural Networks: Full Fine-Tuning, Lightweight FineTuning, and Low-Rank Adaptation

Yizheng Wang, Jinshuai Bai, Mohammad Sadegh Eshaghi, Cosmin Anitescu, Xiaoying Zhuang, Timon Rabczuk, and Yinghua Liu PDF

  • We explore the generalization capability of transfer learning in the strong and energy form of PINNs across different boundary conditions, materials, and geometries. The transfer learning methods we employ include full finetuning, lightweight finetuning, and Low-Rank Adaptation (LoRA).
AMS
sym

Towards the future of physics- and data-guided AI frameworks in computational mechanics

Jinshuai Bai, Yizheng Wang, Hyogu Jeong, Shiyuan Chu, Qingxia Wang, Laith Alzubaidi, Xiaoying Zhuang, Timon Rabczuk, Yi Min Xie, Xi-Qiao Feng, and Yuantong Gu PDF

  • Discuss the future of foundational frameworks that combine the strengths of physics and data.
CMAME
sym

BINN: A deep learning approach for computational mechanics problems based on boundary integral equations

Jia Sun, Yinghua Liu, Yizheng Wang, Zhenhan Yao, and Xiaoping Zheng

PDF

  • Combine boundary element method with PINNs for the first time.
CMAME
sym

A robust radial point interpolation method empowered with neural network solvers (RPIM-NNS) for nonlinear solid mechanics

Jinshuai Bai, Gui-Rong Liu, Timon Rabczuk, Yizheng Wang, Xi-Qiao Feng, YuanTong Gu

PDF

  • We proposed a robust radial point interpolation method empowered with neural network solvers (RPIM-NNS) for solving highly nonlinear solid mechanics problems.
CMAME
sym

Energy-based physics-informed neural network for frictionless contact problems under large deformation

Jinshuai Bai, Zhongya Lin, Yizheng Wang, Jiancong Wen, Yinghua Liu, Timon Rabczuk, Yuantong Gu, and Xi-Qiao Feng

PDF

  • we propose an energy-based physics-informed neural network (PINN) framework for solving frictionless contact problems under large deformation
CMAME
sym

Variational Physics-informed Neural Operator (VINO) for Solving Partial Differential Equations

Mohammad Sadegh Eshaghi, Cosmin Anitescu, Manish Thombre, Yizheng Wang, Xiaoying Zhuang, Timon Rabczuk

PDF

  • We combine PDEs and data using variational principles for solving partial differential equations, and the advantage of the proposed framework is that it can also work without data.
Renewable Energy
sym

Multi-scale modeling in thermal conductivity of Polyurethane incorporated with Phase Change Materials using Physics-Informed Neural Networks

Bokai Liu, Yizheng Wang, Timon Rabczuk, Thomas Olofsson, Weizhuo Lu

PDF

  • We propose a hierarchical multi-scale model RVE utilizing Physics-Informed Neural Networks (PINNs).
Composite Structure
sym

Explainable machine learning for multiscale thermal conductivity modeling in polymer nanocomposites with uncertainty quantification

Bokai Liu, Pengju Liu, Yizheng Wang, Zhenkun Li, Hongqing Lv, Weizhuo Lu, Thomas Olofsson, and Timon Rabczuk

PDF

  • We propose an innovative approach integrating interpretable stochastic machine learning with multiscale analysis to predict the macroscopic thermal conductivity of graphene-based polymer nanocomposites.
Computer & Structures
sym

Energy-based methods for solving forward and inverse linear elasticity problems in 2D structures

Manish Thombre, Cosmin Anitescu, BVSS Bharadwaja, Yizheng Wang, Timon Rabczuk, Alankar Alankar

PDF

  • An additional strong-form loss is enforced on the interface to improve accuracy, incurring only negligible computational overhead in the Deep Energy Method.
Neurocomputing
sym

Applications of scientific machine learning for the analysis of functionally graded porous beams

Mohammad Sadegh Eshaghi, Mostafa Bamdad, Cosmin Anitescu, Yizheng Wang, Xiaoying Zhuang, Timon Rabczuk

PDF

  • We propose a deep learning framework using PINNs, DEM, and Operator learning (FNO: Fourier Nueral Operator) for the analysis of functionally graded (FG) porous beams.

๐Ÿ“ Under Review

EAAI
sym

Artificial intelligence for partial differential equations in computational mechanics: A review

Yizheng Wang, Jinshuai Bai, Zhongya Lin, Qimin Wang, Cosmin Anitescu, Jia Sun, Mohammad Sadegh Eshaghi, Yuantong Gu, Xi-Qiao Feng, Xiaoying Zhuang, Timon Rabczuk, and Yinghua Liu PDF

  • We review AI for PDEs in computational mechanics, including solid mechanics, fluid mechanics, and biomechanics.
IJNME
sym

HomoGenius: a Foundation Model of Homogenization for Rapid Prediction of Effective Mechanical Properties using Neural Operators

Yizheng Wang, Xiang Li, Ziming Yan, Jinshuai Bai, Bokai Liu, Timon Rabczuk, and Yinghua Liu

PDF

  • We propose a numerical homogenization model based on operator learning: HomoGenius. The proposed model can quickly provide homogenization results for arbitrary geometries, materials, and resolutions, increasing the efficiency by a factor of 1000 compared to traditional numerical homogenization methods.
IJMS
sym

Multi-Head Neural Operator for Modelling Interfacial Dynamics

Mohammad Sadegh Eshaghi, Cosmin Anitescu, Navid Valizadeh, Yizheng Wang, Xiaoying Zhuang, Timon Rabczuk

PDF

  • We introduce the Multi-Head Neural Operator (MHNO), a novel neural operator architecture built to handle long temporal dynamics. MHNO uses time-step-specific projections and message-passing-inspired connections to model full time evolution in a single forward pass.

๐ŸŽ– Honors and Awards

  • 2024.10 I participate in (DACOMA-24) and win the best paper award again!
  • 2024.05 Future Scholar at Tsinghua University (ๆธ…ๅŽๅคงๅญฆๆœชๆฅๅญฆ่€…)๏ผ
  • 2022.09 I participate in (DACOMA-22) and win the best paper award!
  • 2022.01 My team placed 18th out of 3537 contestants in the 2021 Kaggle Competition Public Committee (I was the first author)!
  • 2020.06 My team won the 12th place in the 2020 Baidu Star Developer Competition (out of 2,312 teams)!
  • 2019.09 - 2022.06 I participated in the John Ma Cup of Tsinghua University in total 6 times, and won 5 times in the top 8 of the school!

    ๐Ÿ“– Educations

  • 2024.08 - future, Ph.D in Department of Engineering Mechanics, Tsinghua University, Beijing, China
  • 2019.08 - 2022.06, Master in Department of Engineering Mechanics, Tsinghua University, Beijing, China
  • 2012.08 - 2016.06, Bachelor from College of Air Traffic Management, Civil Aviation University of China, Tianjin, China

    ๐Ÿ’ฌ Conference

  • 2025.7, CSTAM-25 ไธญๅ›ฝๅŠ›ๅญฆๅคงไผš, Changsha, China: Yizheng Wang. โ€œAI for PDEs in computational mechanicsโ€ (Oral presentation)

  • 2024.10, DACOMA-24, Beijing, China: Yizheng Wang. โ€œAI for PDEs in computational mechanicsโ€ Won โ€œBest Paper Awardsโ€ (Oral presentation)

  • 2024.04, CCSM2024 ๅ…จๅ›ฝๅ›บไฝ“ๅŠ›ๅญฆๅคงไผš, Nanjing, China: Yizheng Wang, and Yinghua Liu. โ€œAI for PDEs in solid mechanicsโ€. (Oral presentation and host in computaiontal solid mechanics)

  • 2023.08, CCCM2023 ไธญๅ›ฝ่ฎก็ฎ—ๅŠ›ๅญฆๅคงไผš, Dalian, China: Yizheng Wang, and Yinghua Liu. โ€œDeep energy method based on the principle of possible workโ€ Oral Presentation at โ€œArtificial Intelligence and Its Applications in Computational Mechanicsโ€. (Oral presentation)

  • 2023.07, DACOMA-23, Beijing, China: only participate

  • 2023.05, DDCM2023, Dalian, China: Yizheng Wang, and Yinghua Liu. โ€œA deep complementary energy method for solid mechanics using minimum complementary energy principle.โ€ (Oral presentation)

  • 2022.09, DACOMA-22, Beijing, China: Yizheng Wang. โ€œSolving Partial Differential Equations of Solid Mechanics Based on PINN.โ€ Won โ€œBest Paper Awardsโ€ (Masterโ€™s Thesis Supervised by Prof. Yinghua Liu) (Oral presentation)

  • 2022.07, WCCM2022 ไธ–็•Œ่ฎก็ฎ—ๅŠ›ๅญฆๅคงไผš, Yokohama, Japan: Yizheng Wang, and Yinghua Liu. โ€œA Physics-informed Complementary Energy Form in Solid Mechanics.โ€ Presented at Minisymposium MS1716 โ€œData-driven Approaches in Computational Solid Mechanics.โ€ (Oral presentation)

    ๐Ÿ’ป PROFESSIONAL EXPERIENCE

  • 2023.09 - 2024.08, Guest Researcher supervised by Prof. Rabczuk, Weimar, Germany.
  • 2022.11 - 2023.07, Research Assistant at Microsoft Research AI4Science, Beijing, China.
  • 2020.06 - 2020.08, Research Assistant at Shanxi Intelligence Institute of Big Data Technology and Innovation (SIBD) , Taiyuan, China.
  • 2016.07 - 2018.05, Flight Dispatcher at Zhejiang Loong Airlines Co., Ltd., Xiaoshan, China.

Reviewer:

Journal: Journal of the Mechanics and Physics of Solids | Engineering Applications of Artificial Intelligence | Engineering Geology | Underground Space | International Journal of Impact Engineering | International Journal of Hydrology | Engineering Analysis with Boundary Elements | Energy and AI | Neural Network | Frontiers of Structural and Civil Engineering | Computer, Material and Continua | Engineering Structure | International Journal of Mechanical System Dynamics | Scientific Reports | Applied Physics A | International Journal of Mechanics and Materials in Design | Machine Learning for Computational Science and Engineering | Mechanics Based Design of Structures and Machines | International Journal of Hydromechatronics | Computers and Chemical Engineering

Conference: 2025 23rd European Control Conference

My CV

My Recommendation from Timon Rabczuk