Structural Engineering

Permanent URI for this collection

http://etd.aau.edu.et/handle/12345678/138

Browse

Browsing Structural Engineering by Author "Abrham Gebre"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    An Integrated Approach to Sustainable Construction Through Recycling, Co2 Capture and Artificial Intelligence
    (Addis Ababa University, 2026-07-01) Hintsa Gebrezgiher Gebremariam; Abrham Gebre
    The construction industry is one of the largest contributors to climate change, driven largely by carbon dioxide (CO₂) emissions from anthropogenic activities. In addition to putting tremendous strain on natural resources through aggregate extraction, the production of cement and concrete alone is responsible for over 8% of global CO2 emissions. Concurrently, the construction sector produces enormous amounts of construction and demolition (C&D) waste, which exacerbates resource and environmental problems. Recycling this waste into recycled concrete aggregates (RCA) provides a sustainable path forward by conserving raw materials, reducing pressure on landfills, and enabling the sequestration of CO₂. However, RCAs' limited range of applications is due to their lower quality when compared to natural aggregates (NAs). While many studies have explored not only how to enhance RCA performance but also how to effectively incorporate it into construction applications, several important gaps still remain. These include, inconsistencies in results on the influence of parent concrete (PC) strength on RCAs and RAC (recycled aggregate concrete) characteristics, the interaction between carbonation treatment and aggregate characteristics, and the effect of natural preprocessing carbonation on the carbonation potential of the accelerated carbonation. In addition, predictive modelling frameworks capable of linking CO₂ uptake with the mechanical performance of concretes containing carbonated aggregates are still unexplored. This dissertation addresses these gaps through a comprehensive experimental and data driven programs. RCAs were obtained from both laboratory produced concretes of varying strengths and demolished structures. Their physical and mechanical properties were characterized under controlled crushing, ambient carbonation, and pressurized accelerated carbonation conditions. Custom built carbonation chamber with controlled environment and 99.5% purity CO2 gas purchased from local suppliers was used in the pressurized carbonation process. Advanced analytical techniques, including thermogravimetric analysis (TGA-DTA), X-ray diffraction (XRD), and Fourier transform infrared spectroscopy (FTiR), were employed to assess microstructural and chemical changes before and after carbonation. Concrete mixes produced with these aggregates were evaluated for compressive, split tensile, and flexural strength. Complementary regression based machine learning models were developed using 108 datasets to predict compressive strength, explicitly incorporating aggregate quality parameters, mix proportioning parameters, carbonation degree, and percent replacement of natural aggregates by RCAs. The results demonstrated that PC strength has an effect on the characteristics of RCAs and RAC. Higher strength PCs produced aggregates with improved abrasion resistance, impact resistance, crushing resistance, and specific gravity, alongside reduced water absorption, although mortar content increased. Carbonation treatment further enhanced performance, with natural storage achieving up to 7.5% CO₂ uptake by mortar mass and accelerated carbonation achieving up to 12% CO₂ storage in the form of calcium carbonate polymorphs. Concrete produced with carbonated aggregates exhibited strength improvements of up to 12.5% in compression, 8.6% in tension, and 3.3% in flexure. Replacing normal strength aggregates with those from high strength concretes yielded further gains of 18%, 26%, and 30% respectively. Microstructural analyses confirmed that carbonation refined pore structures and improved durability, although dense aggregates from high strength concretes limited CO₂ penetration and resulted in incomplete carbonation. The machine learning models provided robust predictive capability, with ensemble methods such as Random Forest achieving the highest accuracy within a ±13% maximum error ranges. Sensitivity analysis identified mix proportioning, carbonation degree, and aggregate performance as the most influential factors. While the controlled laboratory environment ensured systematic testing, real world demolition waste presents greater variability due to the presence of impurities and weak segregation practices, which remains a limitation of this study for large scale application. In the Ethiopian context, both challenges and opportunities exist for mainstreaming RACs. Establishing national guidelines, improving demolition and waste segregation practices, and integrating CO₂ utilization with industrial emitters such as cement and steel plants will be critical for adoption. With the current trend in corridor development and city renewals in Addis Ababa and other cities of Ethiopia, embedding RACs into the construction sector can establish a circular economy model that reduces reliance on virgin materials, diverts waste from the demolition processes, and supports the country’s climate commitments. To establish waste materials as a sustainable building material and a pillar of low carbon development, legislators, researchers, industry, and contractors must work together.
  • No Thumbnail Available
    Item
    Probabilistic Impact Assessment of Traffic Overload on Bridge Structural Capacity: a Case Study Approach
    (Addis Ababa University, 2025-12-01) Gebyaw Amare; Abrham Gebre
    This study investigates the effects caused by traffic overload on bridge superstructures, especially with regards to heavy vehicles. The critical thresholds with regards to heavy traffic are discussed, taking into consideration parameters such as traffic flow and characteristics and types and patterns of overload. The work also considers design standards and bridge materials. The methodology would commence with identification of some bridges in Ethiopia that are faced with varying degrees of traffic overload. And to get this study off the ground, data was collected from 38,188 truckloads of traffic at weigh stations managed by the Ethiopian Roads Administration in those regions that record high traffic flow of heavy vehicles. Using the R statistical software, a probabilistic loading process was generated. This involved the examination of the material properties of key components as well as the design parameters of a chosen bridge. Probabilistic axle loads were analyzed to assess the impact on the existing bridge’s capacity. A simulation program was developed to determine static overload influence on a particular type of three-span simply-supported box-girder bridge (Mille. 3 bridge). The extreme values of the load effects for various return-period values were computed with statistical extrapolation methods. By using the First Order Reliability Method (FORM), reliability indices (β) can be determined in order to calculate the probability of failure regarding flexural and shear modes. The findings contribute to performance-based management in bridges, considering the impact of real-life overloaded traffic on structural reliability in relation to the life cycle maintenance approach in accordance with AASHTO LRFD/ERA.