Zhang Yuyi - Senior Researcher - Research Center of Dynamics

My research interests focus on the integration of artificial intelligence with solid mechanics and computational mechanics. In particular, I am interested in data-driven and physics-guided approaches for constitutive modeling, elastoplasticity, viscoplasticity, and material behavior prediction. My work explores how machine learning methods, such as neural networks, surrogate models, and physics-informed learning frameworks, can be combined with classical mechanical principles to improve the accuracy, robustness, and interpretability of computational models. I am also interested in inverse problems and parameter identification for constitutive models, especially under conditions involving limited experimental data, parameter non-uniqueness, and strong model uncertainty. Another important direction of my research is the development of neural constitutive models that respect physical constraints, such as thermodynamic consistency, yield conditions, evolution laws, and numerical stability under different loading paths and time discretizations. More broadly, my scientific interests include computational solid mechanics, artificial intelligence for materials modeling, physics-informed machine learning, surrogate-assisted numerical simulation, uncertainty and identifiability analysis, and the application of interpretable AI methods to engineering mechanics problems.