Automated operational modal analysis with adaptive DBSCAN-based algorithm and its engineering application

Wei Liu

doi:10.1007/s13349-025-01002-z

Automated operational modal analysis with adaptive DBSCAN-based algorithm and its engineering application

2025-10-04

0

PRE

AI

DOI：10.1007/s13349-025-01002-z

Original

Original request for help

Save

Abstract

En 中文

In recent years, the interest in robust automated operational modal analysis (AOMA) techniques has increased due to the growing demand for health monitoring of existing heritage buildings. In this study, an AOMA strategy based on the stochastic subspace identification (SSI) technique and the density-based spatial clustering of applications with noise (DBSCAN) algorithm is proposed. This strategy involves the adaptive selection of key parameters for the DBSCAN algorithm and screening of candidate physical clusters. Its highlights include effectively addressing challenges such as closely spaced modes, modal splitting, weak excitation and limited sensor placement (including the number and position of the measured degrees of freedom). The accuracy of the proposed framework is validated through numerical simulations, with its robustness assessed under varying key parameters and the presence of measurement noise. The framework’s effectiveness is demonstrated through ambient excitation tests on the Yingxian wooden pagoda, the oldest and tallest wooden pagoda in China.

Keywords:

Automated operational modal analysis

Stabilization diagram

Density-based spatial clustering of applications with noise (DBSCAN)

Closely spaced mode

Modal splitting

Automated operational modal analysis with adaptive DBSCAN-based algorithm and its engineering application

Abstract

Journal

Organization