Electromechanical impedance based structural health monitoring using extreme learning machine /
Electromechanical Impedance (EMI) based structural health monitoring (SHM) techniques have been developed using a variety of smart material technologies to form a new non-destructive evaluation (NDE) method. In EMI technique, the key indicator of damage is the change in the admittance signature of a...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Kuala Lumpur :
Kulliyyah of Engineering, International Islamic University Malaysia,
2017
|
| Subjects: | |
| Online Access: | Click here to view 1st 24 pages of the thesis. Members can view fulltext at the specified PCs in the library. |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Electromechanical Impedance (EMI) based structural health monitoring (SHM) techniques have been developed using a variety of smart material technologies to form a new non-destructive evaluation (NDE) method. In EMI technique, the key indicator of damage is the change in the admittance signature of a bonded Piezoelectric (PZT) transducer. The purpose of this thesis is to explore the importance and effectiveness of impedance-based structural health monitoring from both hardware and software standpoints. Prior to conducting the experiments, the work first commenced with numerical simulation to test the capabilities and limitations of the Finite Element Method (FEM) in simulating different cases of EMI application. A three-dimensional coupled field finite element model was developed using ANSYS software and the harmonic analysis was conducted for high-frequency impedance analysis procedure. The numerical simulations were carried out to find the effect of different types of damage such as cracks, corrosion, delamination, and other damage on the system's output. The work then followed up with a presentation on EMI parametric investigation, where the effects of different variables such as patch type and size, notch shape and size, glue thickness, patch position were studied. The work then addressed the imperative issue related to EMI's efficiency if both damage and loading co-exist in the structure. The study then explored another issue facing EMI with regards to damage detection and temperature effects since it is a critical problem for structural health monitoring based on electromechanical impedance, especially in detecting low damage levels. An efficient compensatory method for temperature effects was developed. Furthermore, damage identification and location was developed and presented based on extreme learning machine (ELM). The model was trained on simulation-generated data and tested on experiments for estimating the damage location and identification by using piezoelectric sensor data. The numerical results have been validated either experimentally using laboratory equipment or by employing published results available in the open literature and a good agreement has been observed. The practical implementation of the compact EMI method utilized as its main apparatus an HP4194 impedance analyser which served to read the in-situ EMI of piezoelectric sensors attached to the monitored structure. According to the results, use of this method is possible and accurate to monitor the structure's health; even the presence of small damages to the system can be detected. Using damage indicators calculated from the impedance signature of a pristine and damaged beam, it was shown that the EMI technique performs very well for damage detection. As demonstrated through laboratory tests, the results highlight the ability of the machine learning approach to estimating the damage location and identification with good accuracy. Hence, a complete EMI based structural health monitoring system has been developed and can be used for continuous monitoring of structures, thus, enhancing the reliability and accuracy of structural health monitoring systems. |
|---|---|
| Physical Description: | xvii, 177 leaves : illustrations ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 154-162). |