AAU Institutional Repository (AAU-ETD)
Addis Ababa University Institutional repository is an open access repository that collects,preserves, and disseminates scholarly outputs of the university. AAU-ETD archives' collection of master's theses, doctoral dissertations and preprints showcase the wide range of academic research undertaken by AAU students over the course of the University's long history.
How to Submit Your Work
The repository contains scholarly work, both unpublished and published, by current or former AAU faculty, staff, and students, including Works by AAU students as part of their masters, doctoral, or post-doctoral research
- All AAU faculty, staff, and students are invited to submit their work to the repository. Please contact the library at your college.
You may contact digirep@aau.edu.et.with any questions about the repository
Colleges,Institutes in AAU-ETD
Select a college,institute to browse its collections.
Recent Submissions
Automated Classification and Localization of Thoracic Diseases from X- Ray Images Using Deep Learning
(Addis Ababa University, 2025-06) Biruk Tena; Menore Tekeba (PhD)
In developing countries, thorax diseases are the common cause of death. Chest X-ray (CXR) is the most common and effective procedure for the diagnosis of a wide variety of thorax diseases. However, an efficient chest x-ray interpretation requires experienced radiologists, which is particularly lacking in a developing country such as Ethiopia. Efficient computer-aided detection (CAD) of Chest X-ray that can assist radiologists to diagnose lung diseases and offer huge benefits to public healthcare. Currently ChestX-ray14 is the largest public dataset for CXR that includes 14 different pulmonary diseases and one normal class (such as atelectasis, cardiomegaly, consolidation, edema, effusions, emphysema, Fibrosis, Hernia, infiltrations, masses, nodules, no-finding, pneumonia, pneumothorax and Pleural Thickening); This large data set has opened the possibility of creating a better CAD system in radiology. In this study, the proposed Automated Thorax disease Classification and localization (ATDCL) has four main stages, the preprocessing stage, the segmentation stage, the classification stage and the localization stage. Image scaling, normalization, histogram equalization, and median filtering were employed in the preprocessing stage to improve the quality and remove noise from the input images. We have used U-net deep neural network for segmentation. The network composes of two paths: the down-sampling path and the up-sampling path, also known as encoder and decoder. The encoder part used to extract features from the image and then from these latent features, the decoder path will learn to reconstruct the high-resolution binary mask. Then the region of interest was taken out from the chest x-ray image to utilize as input for the next sub stages. In the thorax classification stage, we employed the inception model, a novel deep learning approach. We have achieved the Average accuracy value (Acc) of 0.8489 and we have used a Class Activation Map (CAM) method that can significantly improve the understanding of radiologists about the approximate location of the existed pathology. Our approach can provide radiologists a way to make quicker and more reliable decisions, and providing greater health care system.
Design and Analysis of Boost Converter Sliding Mode Control Using Matlab Real-Time Environment
(Addis Ababa University, 2025-11) Wondimagegn Debebe; Mengesha Mamo (PhD)
Power converters are electronic devices designed to transform and regulate electrical energy for wide range of uses, from small-scale devices like tablets that run on milli-Watts to large-scale power systems to control Mega-watts. There are three main types of DC-DC converters: boost (output voltage higher than input voltage), buck (output voltage lower than input voltage), and buck-boost (either step-up or step-down) converters. In this thesis, two control techniques-proportional-integral-derivative (PID) and sliding mode control (SMC) are used to model and regulate a DC-DC boost converter. We developed a mathematical model of the boost converter using fundamental circuit concepts, resulting in a bilinear system representation. Classical linear control techniques, such as proportional integral- derivative (PID) control, have been extensively employed for boost converter control. However, linear controllers may exhibit limitations when dealing with the nonlinearities, parameter variations, and uncertainties inherent in boost converter systems. To solve these problems, in this study, a sliding mode control (SMC) was designed based on its nonlinear dynamic modeling. This technique has the potential to address the nonlinear nature of the system, provide an improved transient response, and maintain stability across a wide range of operating conditions. The stability analysis of the boost converter was performed using Lyapunov’s stability criterion. The performance of the SMC and PID controllers under various disturbances is thoroughly compared in this work. Across a range of disturbances, the SMC outperformed the PID. measured by figuring out the disturbance Integral Time Absolute Error (ITAE) improvement as a percentage. In particular, it attains an improvement of 97.87% under nominal conditions and 94.86% under modest controller variation. Furthermore, as compared to PID, the SMC showed better robustness in the face of disruptions. By creating Matlab Real-Time Environment (RTE) simulation results using MATLAB software, the efficacy of the suggested control method was confirmed.
Development and Evaluation of Plant-Mediated Nanocomposite Materials for the Disinfection and Removal of Heavy Metals and Fluoride from Drinking Water
(Addis Ababa University, 2025-05) Genet Tsegaye; Zebene Kiflie (Prof.)
Water pollution remains a critical threat to public health and environmental sustainability, especially in low- and middle-income countries where access to efficient and affordable purification technologies is often limited. Conventional treatment systems frequently fail to effectively remove key contaminants such as pathogenic microorganisms, toxic heavy metals, and excess fluoride from drinking water. This study seeks to address these challenges by developing green-synthesized nanocomposite materials, mediated by coffee husk extract (CHE), for advanced water treatment applications including microbial disinfection and the removal of both cationic (Pb²⁺, Cr(VI)) and anionic (F⁻) pollutants. The specific objectives of this study were to: (1) synthesize CHE-capped ZnO nanoparticles (ZnO NPs) for the disinfection of waterborne pathogens; (2) enhance the properties and antibacterial activity of bare CHE-capped ZnO NPs through the incorporation of CHE-capped Fe₃O₄/PU nanocomposites (NCs); (3) develop a CHE-capped magnetite-based pumice silica nanocomposite (CHE-M/PU/Si-NC) for lead ion adsorption; and (4) incorporate CHE-capped MgO NPs and amine functional groups into the CHE-capped M/PU/Si-NC to improve surface charge for the removal of anionic pollutants. For the plant-mediated synthesis of nanomaterials, coffee husk extract (CHE) was obtained using an ethanol-based solid–liquid extraction method. The study focused on optimizing the synthesis parameters of ZnO nanoparticles (ZnO-NPs) by employing CHE as an effective reducing and capping agent to improve nanoparticle size control and functional performance. The optimization involved key parameters such as temperature, zinc precursor-to-CHE ratio, reaction time, and pH. Moreover, the total phenolic content of indigenous CHE, which is essential for understanding its reducing potential, was evaluated and applied throughout all synthesis procedures. The initial formation of the synthesized nanomaterial was visually indicated by a noticeable color change. This was followed by thorough characterization using ultraviolet–visible (UV–Vis) spectroscopy, Fourier-transform infrared spectroscopy (FTIR), and X-ray diffraction (XRD). Further analysis of the physicochemical properties of the synthesized materials was carried out using scanning electron microscopy (SEM), zeta potential analysis, dynamic light scattering (DLS), Brunauer–Emmett–Teller (BET) surface area analysis, and thermogravimetric analysis (TGA).The functional performance of the nanocomposites was assessed through antibacterial activity using agar well diffusion assays, heavy metal quantification via atomic absorption spectroscopy (AAS), and fluoride removal efficiency using a fluoride ion-selective electrode. The optimized CHE-ZnO NPs synthesis conditions included a 1:1 zinc precursor-to-extract ratio, pH 10, reaction temperature of 80 °C, and a 1h reaction time. The resulting ZnO nanoparticles exhibited strong antibacterial activity, against S. aureus and E. coli. Their performance was further improved by forming a composite with Fe₃O₄/PU, yielding particles with enhanced colloidal stability (zeta potential −23.8 mV), reduced size (11.2 nm), and broader-spectrum antibacterial efficacy against both S. aureus and E. coli . For heavy metal adsorption, a green-synthesized CHE capped magnetite-pumice-silica nanocomposite (M/PU/Si-NC) was fabricated and tested for lead removal. The material demonstrated a high surface area (313 m²/g), good thermal stability (up to 690 °C), and a strong negative surface charge (−37.7 mV). It achieved 95% lead removal efficiency at 2 g/L dosage and 100 mg/L Pb²⁺ concentration. Adsorption followed the Langmuir isotherm model with a maximum capacity of 150 mg/g and pseudo-second-order kinetics. The CHE-capped M/PU/Si-NC maintained its initial adsorption capacity after five cycles highlighting its reusability. However, its negatively charged surface limited its ability to capture anionic species such as fluoride and Cr (VI). To address this, the material was modified by incorporating magnesium oxide and amine functional groups, resulting in a positively charged surface under acidic conditions. The amine functionalized magnetite-magnesium silica nanocomposite achieved removal efficiencies of 92% for fluoride and 86% for Cr (VI), the material maintained a relatively high removal efficiency even after multiple cycles. The adsorption behaviors for both pollutants conformed to the Langmuir isotherm and pseudo-second-order kinetics, confirming the efficiency and stability of the modified adsorbent. In conclusion, this research successfully demonstrates the potential of plant-mediated nanotechnology for the development of sustainable water purification materials. The synthesized aforementioned nanocomposites effectively addressed key challenges in water treatment by combining disinfection capabilities with the removal of hazardous ions. Their high performance, environmental compatibility, and reusability make them strong candidates for practical implementation in decentralized or resource-constrained communities, offering a scalable solution to global water quality concerns.
Synthesis of Biodegradable Plastic Film from Waste Sheep Hair keratin and Corn Starch
(Addis Ababa University, 2024-09) Kewani Alemseged; Sintayehu Nibret (PhD)
This study presents the synthesis of a biodegradable plastic film from waste sheep hair keratin and corn starch. Biodegradable plastic films were prepared by first extracting keratin from waste sheep hair and mixing the keratin with corn starch in the starch to keratin ratio of 10:90, 30:70, 50:50, 70:30, and 90:10. The as-prepared biodegradable plastic films were characterized by using X-ray Diffraction (XRD), and UV-Visible spectroscopy. In addition, the effect of operating parameters (processing temperature, processing time and starch to keratin ratios) on the tensile strength and elongation at break of the plastic film were studied. The highest tensile strength (2.37 MPa) was achieved at a processing temperature of 80 °C, a processing time of 30 minutes, and a keratin-to-starch ratio of 30:70. This value is comparable to bioplastics produced from various starch types with glycerol as a plasticizer and without fillers, which typically exhibit tensile strengths ranging from 0.22 to 18.49 MPa. The biodegradability test revealed that the biofilms can be degraded within five days showing the biodegradability potential of bioplastic film from waste sheep hair and corn starch and their future application and capability of replacing fossil fuel-based plastics. The study demonstrates that the prepared bioplastics exhibit good transparency, are safe and environmentally friendly due to their biodegradability, and can be applied in household items, decorations, grocery bags, and related applications.
Seismic Evaluation of Cost-Effective Houses around Addis Ababa Due to Current Earthquakes in Awash Region (Case Study of Condominium Building)
(Addis Ababa University, 2025-09) Suleyman Mohammed; Edom Adane (PhD)
Ethiopia, particularly the Afar region, has experienced a significant increase in seismic activity, including frequent tremors in the Awash region. This heightened seismic risk necessitates a comprehensive evaluation of seismic performance of buildings, especially low-cost housing structures. Seismic evaluation of constructed buildings is mandatory especially in Ethiopia region where the great east African rift valley crosses our country Ethiopia. Studies shows or predicts that the country Ethiopia inclusive of Africa will bisect into two different continents. Recently starting from September to end of October 2024 unusual weekly repeating earthquakes making its initial (epicenter) from awash region is occurring and the ground shake vibration is felt in 10km radius including in Addis Ababa.
The performance of existing building in seismic evaluation area mainly affected by ground excitation. Ground acceleration is the principal cause of collapse in reinforced concrete structures developed in seismically active zone. This might be owing to either the unanticipated nature of the seismic excitation or due to lack of suitable structural design, detailing and construction. It is mandatory to accurately assess the response of the structure during seismic excitation. Codes are unable to foresee the real-time reaction of structures when it comes to response determination. Therefore, the existing structural performance assessment should be done using a reliable and easily applicable analysis procedure.
The revised Ethiopian building code shows a major change on Ethiopia’s hazard map in terms of peak ground acceleration. The expected peak ground acceleration (from 0.05g to 0.1g) for Addis Ababa [1]. And also incorporate two types of spectra that depends the surface wave magnitude (Type 1 and Type 2) that happen recently earth quick fluctuate in those two ranges. Many buildings have been designed and build in accordance with the previous building code for the past few years. This change has the significant impact on these structures’ performance. Therefor it is necessary to evaluate the effectiveness of the current structures using the revised code’s provisions. Because condominiums are built on a large scale across the nation and are places where people congregate, this study focus on them. Premature failures are not anticipated to occur, according to the study's premise. The main objective of the study is to assess the seismic performance of residential condominium buildings that have been built in Addis Ababa for earthquake design. The existing residential condominium buildings have been chosen for the study.From the existing 40/60 high rise condominium houses located in Addis Ababa, four samples (Building Typology of B+G+7, B+G+9,2B+G+12 and 2B+G+18) were chosen for the assessment. Architectural, structural as built structural detail and some design reports were taken from CITY GOVERNMENT OF ADDIS ABABA SAVING HOUSING HOUSE DEVELOPMENT ENTERPRISE for modeling purposes. The structural modeling was analyzed and designed by linear static and nonlinear static pushover analysis using ETABS 22.0.0 (cracked) Engineering software. Furthermore, damage limitation limit state and ultimate limit state have been considered for the linear static analysis.
Base shear, story displacement, story drift, and design spectra were compared in this study in order to analyze the building; The findings base shear, story displacement and story drift significantly changed as the result of revision but spectrum type in the performance evaluation did not significantly change. Lastly, push over analysis was used to evaluate the case study building's performance level. All four typologies used for the study—Building TypologyB+G+7, B+G+9, 2B+G+12, and 2B+G+18)—had sufficient capacity to withstand damage limiting earthquakes, according to the results of linear static analysis. And going additional investigation on nonlinear static analysis showed Only two structures (structure Type B+G+9 and 2B+G+18) had sufficient capacity to ensure life safety but 2B+G+18 has good performance in Y direction between (A and Io), while the other one building (Building Type B+G+7 and 2B+G+12) had inadequate for IO it is in the range of LS(life safety) near Cp (collapse preventive). But the most important discovery was that every type of building did not pass ultimate limit state evaluation, which is a crucial criterion outlined in the code.
The majority of house construction in the town area is from this construction type and the performance of the buildings against earthquake forces was shown to be poor. Accordingly, it is important to seismically retrofit the existing buildings to mitigate potential disasters to be caused by future earthquakes.