07/24/2026
By Marley O'Neil

The Francis College of Engineering, Department of Mechanical Engineering invites you to attend a master's Thesis Mechanical Engineering defense by Callum Hammill titled: "Highly Reduced Order Experimentally Updated Models for Time Response Predictions."

Wednesday, August 5, 2026
11 a.m.-noon
Perry 115

Committee:

  • Advisor: Peter Avitabile, Professor Emeritus, Mechanical and Industrial Engineering, University of Massachusetts Lowell
  • Co-Advisor: Alessandro Sabato, Associate Director, Mechanical and Industrial Engineering, University of Massachusetts Lowell
  • Jesus Reyes Blanco, Associate Teaching Professor, Mechanical and Industrial Engineering, UMass Lowell

Abstract:

Highly reduced order models, enhanced with experimentally measured data, are investigated for use in high-speed computational environments such as shock and blast analysis. These reduced order models are advantageous in terms of computational efficiency but lack the physical elemental characteristics of traditional finite element models. The use of experimental data for updating both reduced order and full space finite element models is investigated in this work to compare the accuracy and performance of each approach. A printed circuit board (PCB) was used as the test article for the development of the full space and reduced space models. Scanning laser vibrometer data was collected to identify the natural frequencies and mode shapes. These data were used for model updating of both the full space and reduced space models using direct and indirect updating techniques. Furthermore, measured impact force and response data were used to validate the updated models through transient response prediction using the Newmark method. Time Response Assurance Criterion (TRAC) values were used to quantify the correlation between the predicted and measured responses, allowing a comparison of the performance of both updating strategies. The developed models, experimental procedures, and updating methodologies are presented and evaluated. Direct and iterative model updating approaches are compared through both modal and transient response validation. Iterative updating is further investigated for damage localization applications. The results demonstrate the capabilities and limitations of reduced order and full space iterative model updating approaches for accurate dynamic response prediction while highlighting the tradeoff between computational efficiency and preservation of physical model characteristics.