Back to search

McMaster University

Data-Driven Signal-Processing and State-Space Tracking Methods for Robust Modal Damping Identification in Dynamical Systems

Abstract

dc:description.abstract

Damping is a crucial characteristic of vibrating systems, as it dictates the rate at which oscillatory energy is dissipated. Accurately estimating damping is therefore essential in structural dynamics and vibration-based structural assessment. While under ideal conditions, damping can often be estimated with precision from an adequately isolated single-mode decay, this task becomes significantly more challenging in field measurements. Contributing factors include sensor noise, finite record lengths, modal overlap, and non-synchronized repeated recordings. This thesis addresses the practical and methodological challenges associated with damping estimation derived from impulse-response measurements conducted under realistic conditions. The central thesis posits that damping estimation is influenced not merely by the final fitting formula but also by the upstream signal-processing decisions that shape the usable decay information. Specifically, the research investigates the alignment and aggregation of repeated measurements, the extraction of modal envelopes, and the modeling of the envelope-recovery problem itself. Three primary contributions are made in this work. First, wavelet-domain alignment and aggregation strategies are developed for repeated impact measurements exhibiting timing offsets, demonstrating that averaging prior to any nonlinear damping extraction process can significantly enhance reliability. Second, a systematic data-driven framework is introduced for designing window/filter-based envelope estimators, revealing that Gaussian-windowed wavelets are not uniformly optimal for the purpose of damping estimation. Finally, a complex state-space framework based on Kalman filtering and Rauch–Tung–Striebel (RTS) smoothing is presented for envelope tracking in both linear and amplitude-dependent nonlinear damping scenarios, indicating that the simpler RTS-Kalman Filter (RTS-KF) formulation excels in robustness compared to the RTS-Extended Kalman Filter (RTS-EKF) in the tested regimes. The thesis concludes that reliable damping estimation necessitates the careful design of the entire inference pipeline, extending beyond the final estimator alone. The proposed methodologies not only provide significant methodological insights but also offer practical guidance for enhancing damping estimation under real-world measurement constraints.

Degree

thesis:*
Department dc:contributor.department
Software Engineering
Year dc:date.issued
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Mohammadi Daniali, Hadi
Advisor dc:contributor.advisor
  • von Mohrenschildt, Martin

Subjects

dc:subject × 3

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivs 2.5 Canada
Language dc:language.iso
en

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
McMaster University
Base URL
macsphere.mcmaster.ca/server/oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Mohammadi Daniali, Hadi. Data-Driven Signal-Processing and State-Space Tracking Methods for Robust Modal Damping Identification in Dynamical Systems. 2026. https://hdl.handle.net/11375/33209