报告人: Nataša Trišović, 教授，贝尔格莱德大学，塞尔维亚

报告人简介：塞尔维亚贝尔格莱德大学机械工程学院教授，主要从事理论与应用力学、工程结构动力学与振动、工程结构可靠性分析等的研究，是CEEPUS (Central European Exchange Program for University Studies 中欧高校学生交流计划) 贝尔格莱德大学分部的负责人。分别担任2019-2024欧盟COST (European Cooperation in Science and Technology) CA18203计划和2022-2026年COST CA21106计划的管理委员会成员。是捷克工程大学力学工程学院客座教授（2009），波西尼亚巴尼亚卢卡大学力学工程学院客座教授（2010，2012），斯洛伐克科技大学力学工程学院客座教授，美国莱斯大学MEMS学院客座教授（2012-2014），EUREKA 计划 和ESPRIT 项目成员，塞尔维亚结构完整性与生命协会成员，塞尔维亚力学学会秘书长在学术方面，发表学术论文60多篇，Google Scholar引用327次，H指数为11。国际化工作包括主持2013-2015年中国-塞尔维亚政府间科技合作项目一项；2015-2017年中国-塞尔维亚政府间例会项目一项；在中国皇冠hg·体育(中国)官方网站大学双创周活动中主讲三次全英文课程"Theory of Oscillations" 、"An Introduction to Structural Optimization and Reliability"、"Foundation of ordinary differential equation".

邀请人：李伟

报告地点：腾讯会议 #428-488-261

报告时间：28号从13点开始，每人两小时

报告题目1：Effect of a Breathing Crack on the Random Vibrations of a Beam

摘要：This study investigates the influence of a breathing crack on the random vibrations of a beam, with special attention to geometrical non-linearities arising from the crack behavior. The system is modeled as bilinear due to the alternating opening and closing of the crack during vibration. Random excitations are considered to simulate realistic loading conditions, complementing common harmonic excitations typically observed in rotating machinery with unbalances. The results provide insights into the dynamic response of cracked structures under stochastic loading, which is crucial for structural health monitoring and early damage detection.

报告题目2：Towards Intelligent Reanalysis in Structural Dynamics

摘要：Predictive reanalysis enables efficient updating of structural responses after minor changes in geometry, material properties, or boundary conditions, without full re-computation. Initially based on finite element methods (FEM), these techniques now increasingly rely on Artificial Intelligence (AI), such as neural networks and ensemble models, to improve speed and adaptability in dynamic analysis. This presentation reviews key developments in predictive reanalysis with a focus on AI integration. A numerical study on a cantilever beam illustrates how changes in cross-section affect natural frequencies, comparing AI-based predictions with classical formulas. Results confirm the potential of AI models for fast, reliable assessment of structural modifications, supporting early-stage design and maintenance. Current challenges and future directions, including data quality, model interpretability, and real-time integration, are also discussed.