Filippoupolitis, Marios
(2020)
Detection of trapped survivors in collapsed reinforced concrete buildings using structure-borne sound transmission.
PhD thesis, University of Liverpool.
|
Text
200995426_Aug2020.pdf - Unspecified Download (14MB) | Preview |
Abstract
Every few years an earthquake of high magnitude occurs around the globe resulting in collapsed structures with people trapped inside them. The seismic research method has the potential to detect physical movement or signals by measuring vibration due to structure – borne sound using seismic sensors. The placement and positioning of these sensors is critical for the successful detection of the survivors and prediction models could be developed and used as a tool to assist in the decision making of the rescue teams. This thesis investigates the potential to use Statistical Energy Analysis (SEA) to model the vibration transmission between structural members of a collapsed reinforced concrete building in order to find trapped survivors. Experimental Modal Analysis (EMA) was carried out to validate finite element models of reinforced concrete beam junctions with surface-to-surface contact conditions which are used to investigate the normal contact stiffness between reinforced concrete beams. It is shown that the contact stiffness follows a lognormal distribution and that the mean value could be used as approximation of the contact stiffness in beam junctions. An ensemble of 30 random beam junctions was generated for Monte Carlo simulations with Finite Element Method (FEM) that allowed Experimental Statistical Energy Analysis (ESEA) to be used to determine Coupling Loss Factors (CLFs) between the two beams. These were compared with CLFs calculated using an analytical model based on a lump spring connector (LSC). It was shown that close agreement (difference within 5 dB) was achieved only for bending waves or torsional waves between FEM ESEA and the analytical model up to the frequency where half the bending or torsional wavelength equalled the longest side of the contact area. When all wave types were combined, reasonable agreement (difference within 10 dB) can be achieved at frequencies below 900 Hz. FEM, SEA path analysis and general SEA matrix solutions were used to estimate vibration transmission in piles of beams using both FEM ESEA CLFs and CLFs from an analytical model of a lump spring connector. It was shown that SEA is more accurate when the number of transmission paths increases and that reasonable agreement is achieved between SEA and FEM up to 700 Hz for the majority of the beams in the piles. Additionally, an ensemble of 30 randomly damaged beam-to-column junctions was generated using a Monte Carlo simulation with FEM. This allows an assessment of ESEA with two or three subsystems to be used to determine the CLFs between the beam and the column considering either only bending modes or the combination of all modes. It is shown that the bending modes are dominating the dynamic response of the junctions over the combination of all the modes and that the uncertainty of predicting the CLFs using FEM ESEA is sufficiently low that it should be feasible to estimate the coupling even when the exact angle between the beam and the column is unknown. In addition, the use of two instead of three subsystems for the junction significantly decreases the number of negative coupling loss factors in FEM ESEA indicating that the two-subsystem model provides a reasonable basis on which to build an SEA model.
| Item Type: | Thesis (PhD) |
|---|---|
| Divisions: | Faculty of Humanities and Social Sciences > School of the Arts |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 04 Sep 2020 11:26 |
| Last Modified: | 18 Jan 2023 23:36 |
| DOI: | 10.17638/03098470 |
| Supervisors: |
|
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3098470 |
Dimensions
Dimensions
