School of Engineering (SEng.) Collections
Permanent URI for this collection
Browse
Recent Submissions
1 - 5 of 314
-
ItemDevelopment of a fluidized bed cookstove for direct combustion of charcoal dust(Makerere University, 2025)Conventional direct combustion and gasifier stoves operate a packed bed combustion technology, which is not favorable for combustion of fine solid biomass such as charcoal fines, not least because of the compact nature of these fuels, which block airflow, leading to incomplete combustion. Conversely, fluidized bed combustion technology would effectively combust fine fuels; however, this is largely deployed of the industrial sector for processes such as heating, drying, separation, power generation, among others, and has never been adopted in cookstove designs despite its high-quality combustion and heat transfer potential. This study, therefore, presents the design, modeling, fabrication, and performance evaluation of a fluidized bed cookstove specifically developed for the direct combustion of charcoal dust. The cookstove employed a bubbling fluidized bed combustion mechanism to enhance fuel-air mixing, hence promoting complete combustion and improving heat transfer. The stove design was a result of mathematical modeling using empirical formulae from previous studies, experimental research, and insights from experts, obtained through the use of structured questionnaires. Computational Fluid Dynamics was employed to simulate both the hydrodynamic behaviour and combustion processes within the reactor, thereby predicting the model performance. A prototype was fabricated and tested with five different charcoal species, i.e., Dichrostachys cinerea, Morus Lactea, Piliostigma thonningii, Combretum molle, and Albizia grandibracteata, following ISO 19867-1:2018 testing protocol. Performance indicators assessed included thermal efficiency, firepower, fuel consumption, CO, CO2, and PM2.5 emissions. The cookstove featured a funnel-shaped combustion chamber with a dense phase region (Ø0.106m × 0.119m), a lean phase region (Ø0.212m × 0.064m), and a total reactor height of 0.182m. The combustion chamber was fabricated from 5mm-thick stainless steel, insulated with 30mm of waste glass wool, and enclosed in 1.5mm-thick mild steel cladding. CFD results indicated dense phase particle concentration with no entrainment and a maximum combustion temperature of 726.8℃. The physical prototype achieved an average high power thermal efficiency of 30%, tier 3 for PM2.5 and CO performance. The results demonstrated that the FBC achieved a thermal and emission performance similar to that of ICS (Improved cookstove).
-
ItemChemical recycling of polyethylene terephthalate wastes into monomers for utilization as binding agents in ceramic floor tiles(Makerere University, 2025)The accumulation of plastic waste, particularly polyethylene terephthalate (PET), presents a growing environmental threat due to its non-biodegradable nature and limited recycling efficiency using conventional methods. This research explores a sustainable solution by chemically recycling PET plastic waste into monomers and integrating them as binding agents in ceramic floor tile production. Two chemical recycling techniques, glycolysis and ammonolysis, were used to depolymerize PET wastes into functional monomers, Bis(2-hydroxyethyl) terephthalate (BHET) and terephthalamide derivatives, respectively. These monomers were then incorporated into clay-based ceramic tile formulations using a fixed clay mass of 200 g and varying monomer volumes from 50 to 90 ml. This study involved the production and testing of both monomer samples and ceramic tile samples. Key properties such as thermal conductivity, water absorption, density, specific gravity, flexural strength, and compressive strength were characterized according to relevant ASTM and ISO standards. To further understand the structural integration and chemical composition of the tiles, Fourier Transform Infrared Spectroscopy (FTIR) and Scanning Electron Microscopy with Energy Dispersive X-ray Spectroscopy (SEM/EDX) were also conducted. FTIR analysis revealed the presence of functional groups such as hydroxyl (-OH), carbonyl (C=O), and ester groups, confirming the successful incorporation of PET-derived monomers into the ceramic matrix. SEM images displayed a more compact and homogenous microstructure in tiles produced using glycolyzed PET monomers, while EDX analysis confirmed the elemental composition, including the distribution of Si, Al, Fe, and traces of carbon indicative of monomer integration. These findings provided insight into the microstructural and molecular interactions between the recycled PET waste monomers and ceramic components. Results indicated that tiles made using glycolysis-based monomers exhibited better mechanical and thermal properties than those from ammonolysis, with compressive strengths reaching 34.25 MPa and flexural strengths exceeding 14 MPa at optimal monomer volume. Water absorption rates declined with increasing monomer content, indicating reduced porosity and improved structural integrity. Both glycolysis and ammonolysis were effective in producing high-performance tiles; however, glycolysis showed superior compatibility with the ceramic matrix due to better polymer dispersion. This work not only demonstrated the feasibility of using recycled PET wastes in ceramic tile production but also contributes to sustainable materials innovation by diverting plastic waste from landfills and reducing reliance on traditional petrochemical-based binder
-
ItemOptimisation of GRID stability in Uganda through vehicle-to-GRID technology(Makerere University, 2025)The growing adoption of Electric Vehicles (EVs) presents both opportunities and challenges for power systems, particularly in developing countries like Uganda, where grid infrastructure is relatively weak. This study investigated the potential of Vehicle-to-Grid (V2G) technology to optimize grid stability in Uganda by addressing voltage deviation and peak demand issues. A quantitative approach was employed, combining empirical data collection with simulation-based modeling and statistical analysis. Charging data from Kayoola electric buses with real-world load profiles from the 55 MVA Kisugu substation in Kampala were analyzed to assess baseline grid stress. Simulations were performed using MATLAB/Simulink (R2024a), complemented by statistical correlation analysis, probabilistic modeling with the Weibull distribution, and optimization through a rule-based control algorithm. The results demonstrate a statistically significant correlation between EV charging and peak demand (r = 0.68, p < 0.05), confirming that EV charging can contribute to grid stress. Voltage drop simulations revealed that under peak demand conditions, deviations exceeded Uganda’s 5% regulatory threshold, reaching as high as 10.4% in the assessed daily load profile. With V2G integration, voltage drops were significantly reduced, falling within acceptable limits across the intervals. The optimal V2G strategy was based on predefined voltage deviation thresholds and real-time control of EV discharge, factoring in fleet size and State of Charge (SOC). For the daily load profile considered, a coordinated fleet of 54 EVs, each maintaining a minimum SOC of 73.33%, collectively was required to supply up to 14.72 MW of power to stabilize the grid during peak demand. The study concludes that V2G technology offers a technically viable and contextually relevant solution for improving grid resilience in Uganda. Recommendations are provided for improving access to high-resolution grid data, piloting V2G operations, and exploring policy frameworks that align with Uganda’s operational realities. The findings contribute a localized and data-driven reference model for V2G deployment in Sub-Saharan Africa, helping to fill a critical gap in the global literature on EV-grid integration.
-
ItemDevelopment of a hydropower forecasting model at Isimba HPP on River Nile(Makerere University, 2025)This study compared the performance of Multi-Layer Perceptron Feed Forward Neural Network (MLP-FFNN), Grey model, Convolutional Neural Network (CNN) and the Random Forest (RF) regressor for development of a hydropower prediction model for the 183 MW Isimba Hydropower Plant. In addition, a physics based model was obtained from the standard hydropower generation potential equations was implemented as a baseline for comparison. Prediction models based on hourly data were developed; the physics based model was used as a baseline for comparison, the RF, MLP-FNNN and CNN models for prediction of output power generation. The physics based model performed the well with NSE of 0.68 and very minimal errors MAE of 10.711 and RMSE of 14.6161. The Grey Model showed moderate improvement NSE of 0.3145, MAE of 14.6161. Among machine learning methods, the MLP-FFNN achieved the best performance with NSE of 0.752 and the lowest errors MAE of 8.72, RMSE of 13.53. Random Forest also gave good results NSE of 0.5997, RMSE of 11.713 while CNN attained the NSE 0.731 but higher errors MAE of 8.3659, RMSE of 11.6835. These results confirm that the MLP-FFNN outperformed both RF and CNN, making it the most accurate model. The supervised deep-learning model developed for the prediction of hydropower generation will have the potential to reduce on the operational and maintenance costs and increase or optimize the energy output of hydropower generation. The results can thus help policymakers and organizations to plan energy management using evidence-based forecasts and manage water and energy resources more efficiently.
-
ItemGeometric and hydraulic parameter correlation with cross pipe culvert extension requirements in road rehabilitation works( 2025-11-11)Highway rehabilitation often results in modifications to the existing embankment geometry and roadway prism, necessitating the extension of cross-pipe culverts to maintain hydraulic and geometric continuity. These extensions are commonly determined through subjective judgment or rule-of-thumb methods, causing design inconsistencies, project delays, and cost overruns. The absence of scientifically established literature guiding the quantification of culvert extension lengths limits a replicable and consistent approaches for extension length determination. This study develops and validates a predictive, data-driven framework that integrates Digital Terrain Model (DTM)–derived parameters with analytical and hydraulic principles to quantify culvert extension lengths and improve design-stage consistency. Using the Alwii–Nebbi corridor (Uganda) as a case study, Digital Terrain Models (DTMs) derived from topographic surveys were integrated with roadway geometry to extract governing variables—shoulder elevation (Z), invert elevation (V), lateral offsets (Xz, Xv), embankment slope (S), culvert gradient (g), diameter (Ø), and wall thickness (t)—across 61 culvert sites (122 inlet/outlet cases). From the roadway–embankment–culvert cross-section, explicit closed-form expressions for upstream (Lu) and downstream (Ld) extensions were derived, with √(〖(g〗^2+1)) capturing barrel inclination and (S ± g) distinguishing inlet/outlet configurations. Terrain analysis revealed corridor variability (Z−V ≈ 2.0–3.5 m; Xv ≈ 5.0–6.5 m) and diagnostic negative extensions indicating regrading needs rather than pipe addition. Validation showed strong agreement between analytical predictions and DTM-measured lengths (R² > 0.9; low RMSE) and hydraulic adequacy under design flows (Hw/D ≤ 1.2 via HY-8). Sensitivity and correlation analyses identified vertical separation ΔZV = Z − (V + Ø + t) and lateral placement (Xv − Xz) as dominant controls; Xv offers the most actionable field lever for fine-tuning lengths. Descriptive statistics indicated a modest downstream bias (mean Ld ≈ 1.64 m vs. Lu ≈ 1.45 m; Δ ≈ +0.19 m) with typical extensions between 1.0 and 2.0 m, supporting modular sizing (1.0–1.5–2.0 m) while retaining site-specific optimization. The study contributes a reproducible, terrain-sensitive, and hydraulically verified framework suitable for specification adoption. It recommends institutionalizing the DTM → analytical sizing → HY-8 check workflow in design manuals and Terms of Reference (ToRs). Future research should focus on evaluating the structural performance of culvert extension joints and joint treatment during installation, and on integrating the developed equations into civil engineering platforms such as AutoCAD Civil 3D, ArcGIS, and HY-8 to enable automated optimization in culvert extension planning.