Nazarbayev University Repository (NUR) is an institutional electronic archive designed for the long-term preservation, aggregation, and dissemination of scientific research outcomes and intellectual property produced by the Nazarbayev University community and affiliated organizations.
NU Research Portal - here you can find information about research projects, publications, and other activities of our 400+ researchers and faculty members.
results
Discover
Communities in DSpace
Select a community to browse its collections.
Recent Submissions
Item type:Item, Access status: Embargo , DEVELOPMENT OF CENTRIFUGAL ADDITIVE MANUFACTURING TECHNOLOGIES USING FINE POWDER(Nazarbayev University School of Engineering and Digital Sciences, 2025-08-15) Berkinova, ZhaziraPowder-based Additive Manufacturing (AM) processes offer the potential to fabricate complex 3D components with high precision. The quality and performance of printed components are strongly influenced by powder characteristics, laser operating parameters, and the thermal behavior of a powder bed during the melting process. Specifically, powder characteristics such as size, shape, and size distribution play a critical role by affecting powder flowability, spreadability, and compactability. Due to the high cost of producing uniformly sized spherical metal powders, commercially available feedstocks often exhibit varied sizes and shapes. This makes powder mixing a critical pre-processing step in AM to prevent particle segregation and ensure uniform layer deposition. However, limited understanding exists regarding the flow and mixing behavior of irregularly shaped particles. To address this, the first part of the thesis investigated the mixing behavior and flowability of differently shaped powders in a rotating drum using experiments and Discrete Element Method (DEM) simulations at various rotational speeds. The obtained results demonstrated that the mixing rate constant increased notably from 10 to 50 rpm, with only slight improvement at 60 rpm. Beyond 60 rpm, mixing efficiency declined due to a shift in flow regimes. Despite a minor mixing rate gain at 60 rpm, the highest packing fraction was observed at 50 rpm, identifying it as the optimal speed for effective mixing of differently shaped powders. Powder size and geometry have a strong influence on powder distribution and packing fraction within the layer in additive manufacturing. Fine powders (d50<20 μm) offer improved melt track stability and fabricated component precision. However, fine particles often exhibit poor flowability and spreadability due to their high surface area and cohesive forces, leading to agglomeration and uneven spreading. To address these issues, the second part of the thesis introduced a novel compaction method for fine particles using centrifuge-assisted artificial high gravity. The fine powder bed compaction was analyzed employing lab-scale and large-scale centrifuge machines along with DEM simulations. The results demonstrated that artificial gravity significantly improves fine powder bed compaction in AM. DEM simulations showed an 82.8% rise in packing fraction at 71.7G compared to 1G, with rapid densification occurring within the first two rotations. Moreover, the components fabricated via laser melting under artificial high gravity exhibited reduced defects, confirming the benefits of this compaction method. Furthermore, this PhD thesis analyzed the temperature distribution within the melt pool using the newly developed analytical approach based on a disk-shaped heat source by integrating both point and doublet sources. The proposed approach incorporated heat losses due to conduction, convection, and radiation, in addition to simulating the Marangoni effect. Model validation against numerical data demonstrated excellent predictive accuracy, with over 99% agreement for the peak temperature at the top surface of the AlSi10Mg powder bed. Furthermore, validation against experimental data confirmed the reliability of the model, yielding melt pool width and depth accuracies of 94.6% and 88.1% for AlSi10Mg, and 94.5% and 85.3% for Inconel 625, respectively. Parametric studies reveal that increasing laser power from 150 W to 200 W significantly enlarges the melt pool, with maximum depth rising from 22 µm to 32 µm, indicating full powder bed penetration, thus enhancing the interlayer bonding. The final part examined how artificial gravity influences melt pool formation during laser melting. Volume of fluid simulations revealed that improved powder bed compaction achieved under high gravity conditions resulted in a 25.6% increase in melt pool depth, accompanied by a reduction in melt pool width. At elevated gravity levels, fluid motion shifted from primarily horizontal flow to a downward vertical direction, enhancing the depth of molten material penetration while limiting its lateral expansion. Under 1G, poor compaction led to void-induced balling and surface irregularities. In contrast, artificial high gravity smoothed the melt pool’s top surface. This thesis explores using gravitational acceleration to address compaction challenges with fine and potentially ultra-fine particles, thus expanding their applicability in additive manufacturing. This approach also provides a promising strategy to overcome microgravity limitations, making additive manufacturing processes feasible in space.Item type:Item, Access status: Open Access , A Systematic YOLO-Specific Model Selection for Mechanical Fault Identification in High-Voltage Insulators(Nazarbayev University School of Engineering and Digital Sciences, 2025) Serikbay, Arailym; Nurmanova, Venera; Akhmetov, Yerbol; Zollanvari, Amin; Bagheri, MehdiRegular monitoring of outdoor insulators is crucial to ensure the reliable functioning of the power grid. With recent progress in computer vision technologies, traditional manual and expensive visual inspections can now be replaced by automated analysis using images captured by unmanned aerial vehicles (UAVs). In such applications, a practitioner might opt to choose a state-of-the-art object detection and classification deep learning architecture, including You Look Only Once (YOLO). Te variety of existing YOLO architectures per se makes selecting the best application-dependent YOLO model challenging. However, selecting the best architecture solely based on performance without considering the model complexity limits its deployment on resource-limited embedded devices. Consequently, we conduct a rigorous, systematic model selection based on the performance–complexity trade-of across 13 YOLO architectures to determine the most effective model for detecting common mechanical faults in insulators using images captured by UAVs. A dataset comprising 15,000 images of insulators, categorized into normal condition, bird-pecking damage, cracks, and missing caps, has been compiled for training the models. Specifically, all considered YOLO architectures are compared using model complexity and the mAP@0.5:0.95. During the model selection stage, YOLOv8l proved to be the best model in terms of mAP@0.5:0.95, while YOLOv5n was the model of choice in terms of complexity at the expense of a slight reduction in performance. Alongside YOLOv8l and YOLOv5n, an “optimal” model (OP-YOLO) was selected using a multicriteria decision-making approach, balancing detection accuracy and computational efficiency. In particular, in terms of test-set performance,YOLOv8l, YOLOv5n, and OP-YOLO achieved 0.919, 0.901, and 0.896 mAP@0.5:0.95, respectively. Although YOLOv8l reported a higher mAP@0.5:0.95, YOLOv5n requires ∼20.9 times less memory and ∼40.2 times less foating-point operations per second(FLOPs). Also, YOLOv5n outperforms the OP-YOLO model, still requiring ∼12 times less memory and ∼19 times less FLOPsItem type:Item, Access status: Open Access , Energy Generation and Carbon Footprint under FutureProjections (2022–2100) of Central Asian Temperature Extremes(Wiley, 2024) Broomandi, Parya; Bagheri, Mehdi; Fard, Ali Mozhdehi; Fathian, Aram; Abdoli, Mohammad; Roshani, Adib; Shafiei, Sadjad; Leuchner, Michael; Kim, Jong RyeolLimiting the global temperature rise to 1.5 °C is becoming increasingly difficult. The study analyzed data from 700 locations (1962–2100) to assess climate change impacts on heating-cooling energy and carbon footprint in under-researched Central Asia (CA). Under SSP2-4.5, icing and frost days reduce, while summer days and tropical nights increase. Central Asian countries will see an increase in cooling needs despite the projected decline in heating demands, with Kyrgyzstan experiencing the highest rise in cooling degree days, projected to increase by 132% and 165% in the near-future underSSP2-4.5 and SSP5-8.5, respectively. As a result, cooling energy generation is expected to rise by 39% and 92% under SSP2-4.5 and SSP5-8.5, respectively. However, CO2 emissions for cooling are much lower in Kyrgyzstan and Tajikistan due to their reliance on renewable energy. CO2 emissions in these countries are projected to be ≈10 times lower than in other parts of CA. From2022 to 2100, cooling-related emissions are estimated to increase by 41% and80% under SSP2-4.5 and SSP5-8.5, respectively across CA. Urgent adaptation is needed for resilient cities and stable power by expanding renewables, modernizing infrastructure, boosting efficiency, adopting policies, and fostering cooperationItem type:Item, Access status: Open Access , ANALYSIS FOR A POTENTIAL ENHANCEMENT OF BLASTING PERFORMANCE FOR ENSURING GROUND STABILITY IN THE OIL SHALE MINE(Nazarbayev University School of Mining and Geosciences, 2025-07-20) Abdykarimov, ChingizThis thesis investigates how simulation-based blast design enhancements can improve blasting performance. The work builds on the field findings reported by Sabanov et al. (2023). The author created a complete method that mixes advanced models with an uncertainty analysis. The study uses JKSimBlast software to simulate three dimensional blasts, it includes mapping how explosive energy spreads, which relies on Kleine field theory. Monte Carlo simulations with 10,000 trials determine how much uncertainty there is in when a blast starts, how strong the rock is, and how much charge is present. This process predicts blast outcomes and includes the probability of those outcomes. A regression study of sensitivities, shown with tornado charts, points out the most important factors that control how well a blast works - these tools help design the best blast pattern for the mine's complex layers of rock. The pattern expands on previous designs based on experience to handle harder situations. The study's results show the new contributions of this simulation work. The radius of blast damage, which is how much the rock breaks, acts as a random factor - this radius follows a log normal spread, it averages about 0.29 m, which means that cracks do not spread in a fixed, set way, but they vary. The amount of explosive needed per volume of rock changes based on the orientation of the blast hole. Holes in the roof need about 2.02 g/dm3. Floor holes use about 2.12 g/dm3, and wall holes use about 2.49 g/dm3 on average. These differences show how geometric confinement and available free surfaces affect how well blasting works. The sensitivity study shows that the rock's uniaxial compressive strength mainly controls both the blast damage radius also the amount of explosive needed. Borehole diameter comes next. The properties of the explosive, such as its density and velocity of detonation, and factors based on experience, only have small effects. By measuring how much each factor influences the outcome, the study tells practitioners where to focus their efforts to control measures. In conclusion, this research extends the Sabanov’s field results by offering a simulation-based pre-validation framework for blast design. The study shows that a well-planned charge distribution, checked through simulations, can break rock evenly as well as keep the ground stable in different conditions. Engineers can fine tune blast patterns with confidence limits and safety factors. That makes sure that even the worst-case situations stay within safe design limits, this forward-looking, model-based approach improves mine design. It makes blasting operations safer and more effective before full-scale implementation.Item type:Item, Access status: Embargo , Wettability Alteration During Co-current and Counter-current Spontaneous Imbibition of Low Salinity Water in Naturally Fractured Reservoir(Nazarbayev University School of Mining and Geosciences, 2025-09-14) Karimova, MarzhanNaturally fractured reservoirs (NFRs) present significant challenges for fluid flow characterization due to their inherent heterogeneity and complex fracture-matrix interactions. A key mechanism governing fluid exchange in such systems is spontaneous imbibition (SI) - the capillary-driven movement of fluids from fractures into the porous matrix. This process manifests in two dominant modes: co-current spontaneous imbibition (COSI) and counter-current spontaneous imbibition (COUSI), determined by the direction of fluid movement relative to boundary conditions. Analytical descriptions of these imbibition processes rely on diffusivity coefficients that combine relative permeability and capillary pressure parameters. Solutions to the resulting equations can be obtained through perturbation techniques. To validate these models, this study compares numerical predictions with experimentally observed in-situ water saturation profiles, acquired through X-ray computed tomography (CT) imaging. The validation has been performed using spontaneous imbibition tests of co-current and counter-current flows in air-brine system for diatomite rock. The validation showed a good agreement between experimental and numerical data, achieving an average relative error of approximately 10%. The study investigates low-salinity water (LSW) injection as a method to alter wettability toward more water-wet conditions and enhance recovery. Oil-saturated carbonate cores were subjected to imbibition experiments in Amott cells using both high salinity water (HSW) and LSW. Six carbonate core samples were analyzed under controlled boundary conditions, with water saturation profiles monitored at multiple time steps using CT scanning. Experimental results reveal a pronounced improvement in water uptake and imbibition depth under LSW conditions. Cores exposed to low salinity brine exhibited higher and more uniformly distributed water saturations, reflecting enhanced wettability alteration and deeper fluid penetration. Conversely, HSW tests showed lower saturation levels with sharper fronts, indicating limited wettability change and reduced displacement efficiency. These trends were corroborated by numerical simulations, which aligned closely with experimental observations, particularly in LSW scenarios. The research introduces an integrated experimental and numerical modeling approach to comprehensively examine wettability alteration mechanisms during SI in oil-wet carbonate reservoirs. Unlike conventional methodologies focusing merely on final oil recoveries or endpoint saturations, this study employs high-resolution X-ray CT scanning to dynamically visualize and quantify evolving water saturation profiles during both COSI and COUSI. A novel comparative analysis between LSW and HSW imbibition processes highlights substantial differences in saturation front propagation and fluid distribution patterns attributable to dynamic wettability alteration that can be modeled by adjusting key parameters such as relative permeability and capillary pressure throughout the imbibition process. This combined experimental-numerical approach provides unprecedented detail into the temporal and spatial dimensions of wettability alteration induced by LSW. Ultimately, this research delivers critical insights and proposes a robust framework for evaluating and optimizing low-salinity EOR strategies, significantly enhancing predictive capabilities and operational outcomes in fractured carbonate reservoirs.