Advisory Committee Chair
Muhammad Sherif
Advisory Committee Members
Abdullahi Salman
Christopher Waldron
Jeffrey Morris
Nasim Uddin
Document Type
Dissertation
Date of Award
1-1-2025
Degree Name by School
Doctor of Philosophy (PhD) School of Engineering
Abstract
Concrete is the most utilized construction material globally and is the second most consumed material, besides water. It has been used to construct key infrastructures such as bridges that link cities and states, dams for hydroelectric power generation, pavements for connecting urban and rural areas, and commercial/residential buildings that house humans. It has been employed in developing marine, environmental, and coastal defense structures for civilian and military purposes. This diverse array of uses expresses the crucial importance of concrete. However, despite its importance and versatility, concrete is plagued by low tensile strength and high susceptibility to cracking. In marine environments, these cracks have been reported to reduce the durability of concrete due to the ingress of harmful chemical substances, costing millions of dollars in repair and maintenance. If not managed properly, these limitations could result in large-scale disasters. Hence, advanced structural techniques and materials are needed to monitor and mitigate the challenges associated with traditional concrete cracking. This research aims to address the cracking problems associated with traditional concrete by (1) developing Artificial Intelligence (AI)-based Structural Health Monitoring techniques (SHM) for detecting and monitoring cracks, and (2) characterizing advanced cementitious materials with crack-healing abilities. High-level unmanned aerial vehicle-machine learning-based (UAV-ML) solutions were engineered to detect pavement cracks, quantify them, and evaluate their morphological characteristics. Additionally, crack detection and propagation prediction models were developed for Engineered Cementitious Composite structures, providing a preemptive SHM framework for mitigating cracks. Similarly, structural deterioration models to track structural displacement and strain under loading conditions were created, all a part of the designed SHM framework. Likewise, bacteria-based self-healing cementitious composites with high-performance abilities were formulated and experimentally characterized. Based on the science of microbial-induced calcite precipitation, the developed composites were discovered to heal cracks wider than 2.5mm. The flexural and compressive material response of the bacteria-based composites was indicative of their high mechanical performance in addition to their superior self-healing abilities. The structural health monitoring framework, in addition to the bacteria-based composite results, provides a comprehensive body of engineering solutions for tackling concrete cracking on two different levels.
Recommended Citation
Amieghemen, Goodnews Eronmonsele, "Development And Characterization Of Self-Healing Engineered Cementitious Composites Using Computer Vision And Machine Learning Algorithms" (2025). All ETDs from UAB. 6818.
https://digitalcommons.library.uab.edu/etd-collection/6818
Comments
etdadmin_upload_1144183