A Novel Drive-by System Identification Approach for Bridges Utilizing a Modal FRF Similarity Criterion and Soft-Imputing

EasyChair Preprint no. 10079

Abstract

This research presents a novel method for identifying the mode shapes of a bridge using sensors mounted on vehicles crossing the bridge. The proposed technique employs an enhanced soft-imputing matrix completion approach to predict the bridge's vibration signals at virtual fixed points in invalid regions of the response. To improve the accuracy and automate the implementation of the soft-imputing technique, a novel paradigm is developed to find the optimal mode shapes by utilizing the full-field responses of the bridge. The modal frequency re-sponse functions (FRFs) of the bridge are determined using a rational procedure, and the algorithm aims to find the mode shapes in a way that maximizes the simi-larity between the determined modal FRF of the bridge and that of an ideal single-degree of freedom (SDOF) system. Numerical experiments using synthetic data on a one-span bridge were conducted to verify the effectiveness of the proposed algorithm. The results indicate that the method finds the best mode shape from the matrix completion approach without the need for manual and engineering judg-ment, and the main mode shapes of the bridge can be identified accurately.

Keyphrases: Bridge Modal FRF, Indirect Bridge Monitoring, matrix completion, Mode Shape Identification

@Booklet{EasyChair:10079,
  author = {Mohammad Talebi Kalaleh and Qipei Mei},
  title = {A Novel Drive-by System Identification Approach for Bridges Utilizing a Modal FRF Similarity Criterion and Soft-Imputing},
  howpublished = {EasyChair Preprint no. 10079},

  year = {EasyChair, 2023}}