Data-Physics Coupled Driven Structural Design and Health Monitoring Methods

Speaker: XU Yongjia

Venue: LTN-A 301, Haining Campus

Abstract: Intelligent desian and structural health monitoring of civil engineering structures are critical research fields atpresent. This presentation reports the recent progress in solving problems such as low utlization efticiency of existing design data and unstable quality in structural design by introducing generative intelligent design methods.Leveraging advanced generative Al algorithms the proposed methodologies extract key features and patterns from existing design data to acquire engineering structural design capabilities, fulfilling new design tasks efficiently while maintaining stable design quality. Additionallv, the civil enaineering tield has beaun transitioning from aconstruction-tocusedapproach toward emphasizino balanced construction and maintenance.To ensure the satety and functionality of existing structures, structural health monitoring has increasinaly gained importance. This presentation, therefore, introduces a Bidirectional Granhics-Based Diaital Twin (Bi-GBDT) structural health monitorna tramework, Supported by various technologies such as digtal twin, virtual environments, deep learning, and finite element analyss, the Bi-GBDT tramework integrates the data acquisition, external damage identification, and intrinsic damage evaluation to achieve accurate and efficient structural condition assessments.