DSpace Repository :: Browsing 02. Master's Thesis by Title

02. Master's Thesis

Permanent URI for this collection

https://nur.nu.edu.kz/handle/123456789/8136

Browse

Now showing 1 - 20 of 505

Open Access
2D SKELETON-BASED HUMAN ACTION RECOGNITION USING ACTION-SNIPPET REPRESENTATION AND DEEP SEQUENTIAL NEURAL NETWORK
(Nazarbayev University School of Engineering and Digital Sciences, 2022-05) Askar, Aizada
Human action recognition is one of the crucial and important tasks in data science. It aims to understand human behavior and assign a label on performed action and has diverse applications. Domains, where this application is used, includes visual surveillance, human–computer interaction and video retrieval. Hence, discriminating human actions is a challenging problem with a lot of issues like motion performance, occlusions and dynamic background, and different data representations. There are many researches that explore various types of approaches for human action recognition. In this work we propose advanced geometric features and adequate deep sequential neural networks (DSNN) for 2D skeleton-based HAR. The 2D skeleton data used in this project are extracted from RGB video sequences, allowing the use of the proposed model to enrich contextual information. The 2D skeleton joint coordinates of the human are used to capture the spatial and temporal relationship between poses. We employ BiLSTM and Transformer models to classify human actions as they are capable of concurrently modeling spatial relationships between geometric characteristics of different body parts.
Open Access
A 3d multidisciplinary automated design optimization toolbox for wind turbine blades based on ns solver and experimental data
(Nazarbayev University School of Engineering and Digital Sciences, 2018) Sagimbayev, Sagi
This thesis attempts to develop a framework to optimize wind turbine blades automatically by a multidisciplinary 3D modeling and simulation methods. The original NREL Phase VI wind turbine blade and its experimental measurements are used to validate the Computational Fluid Dynamics (CFD) model developed in ANSYS Fluent and based on the 3D Navier-Stokes (NS) solver with a realizable k-epsilon turbulence model, which is later used in the automation process. The automated design optimization process involves multiple modeling and simulation methods using Solidworks and ANSYS Mesher and ANSYS Fluent NS solver, which are integrated and controlled through Matlab by implementing the scripting capabilities of each software package. Then all scripts are integrated into one optimization cycle, with its optimization objective being the highest mean value of 3D Lift/Drag ratio (3DLDR) across the blade. A 3DLDR distribution across the blade can be calculated by the Inverse Blade Element Momentum (IBEM) Method based on experimental measurements. The optimization process is performed to find optimized twist angles across the blade using the Angle of Attack (AOA) with the highest 3DLDR as a reference, in order to 3 achieve the optimization objective. Therefore, the automatic optimization framework is based on 3D solid modeling and 3D aerodynamic simulation and guided by IBEM and experimental data. Thus the design tool is capable of exploiting the 3D stall delay of blades designed by the traditional 2D BEM method to enhance their performances. It is found that this automated framework can result in optimized blade geometries with the improvement of performance parameters compared to the original ones.
Open Access
3D PRINTING OF BIOCOMPATIBLE CRYOGELS FOR BONE TISSUE ENGINEERING
(School of Engineering and Digital Sciences, 2023) Moazzam, Muhammad
Natural biopolymers are highly valued and commonly utilized in tissue engineering to create scaffolds that support living cells. This is due to their exceptional biocompatibility and the fact that their degradation rate can be controlled. However, the shape and average pore size are crucial in biological processes that influence the kinetics of cell proliferation and tissue regeneration processes linked to the production of extracellular matrix. For the construction of high-accuracy hydrogel scaffolds via 3D printing, the shear thinning characteristics of the bioinks used frequently result in morphological compromises like smaller pore diameters. Here, we introduced a new mixture of gelatin and oxidized alginate (Gel/OxAlg) that has been optimized for use in 3D printing and cryogelation techniques. This composite formulation allows for the creation of highly porous and biocompatible hydrogel scaffolds with extra-large pore sizes (d > 100 μm) using a combination of 3D printing and cryogelation techniques. These scaffolds have the potential to serve as a platform for various tissue engineering applications, and their morphological properties and cell viability data can be tailored accordingly. Overall, our approach offers a simple and cost-effective method for constructing hydrogel scaffolds with high accuracy.
Restricted
3D PRINTING OF GELATIN/OXIDIZED CARBOXYMETHYL CELLULOSE SCAFFOLDS WITH GRADIENT POROSITY FOR BONE TISSUE REGENERATION APPLICATIONS
(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-23) Dyussenbinov, Aibek
This master thesis investigates the development and evaluation of 3D-printed gelatin/oxidized carboxymethyl cellulose (OxCMC) scaffolds with gradient porosity for applications in bone tissue regeneration. Recognizing the limitations of current bone repair methodologies, this research aims to mimic the natural extracellular matrix of bone through advanced scaffold engineering techniques. The thesis explores the synthesis and optimization of bioinks from gelatin and OxCMC, chosen for their biocompatibility, biodegradability, and mechanical properties conducive to 3D printing. Through extensive experimentation, including rheological tests, Fourier-transform infrared spectroscopy (FTIR) analysis, and scanning electron microscopy (SEM) imaging, scaffold formulations were tailored to achieve desired porosity gradients and mechanical strength. The novel approach of utilizing a complex 3D printing model with different pinheads for varying ink compositions is highlighted as a key innovation. This method allowed for the creation of scaffolds that not only support cell adhesion and proliferation but also replicate the porosity gradient inherent to natural bone, thereby addressing a critical aspect of scaffold design in bone tissue engineering. Results indicated a direct correlation between the polymer content in the scaffolds and their swelling ability, degradation rates, and mechanical properties. Scaffolds with higher polymer content showed less swelling but greater mechanical strength, aligning with the requirements for supporting bone tissue regeneration. The gradient scaffold, in particular, demonstrated a balance between swelling behavior and mechanical integrity, suggesting its suitability for bone tissue engineering applications. This research contributes to the field of regenerative medicine by offering a promising scaffold design strategy for bone tissue regeneration. By closely mimicking the structural and mechanical properties of natural bone, the developed scaffolds hold potential for improving the outcomes of bone repair and regeneration procedures, paving the way for future clinical applications.
Open Access
3D-PRINTED OSTEOCHONDRAL GRAFTS AND THEIR CHARACTERIZATION
(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-25) Effanga, Victoria Effiong
The osteochondral (OC) interface is a complex tissue with a hierarchical structure found at the ends of the bones of the knee joint consisting of a layer of soft tissue (cartilage) overlaying hard tissue in the subchondral bone. It exhibits a gradient of its constituents, especially in terms of mineral concentration, cell phenotype, collagens, and glycosaminoglycans, with a thickness of around 0.5 mm. The tidemark, a critical yet often overlooked component of OC interface tissue, plays a pivotal role in maintaining tissue function by acting as a barrier against vascular invasion of the cartilage. Fabricating scaffolds that mimic the complex physiology and functionalities of the OC tissue within the physiological thickness remains a challenge. This study aimed at fabricating a unitary composite scaffold that is similar of the OC interface in terms of distribution of its mineral content. It was hypothesized that the interface formed between the layers of the multilayer graft will possess a thickness of hydroxyapatite (HAP) gradient similar to that seen at the native rabbit OC tissue. To test the hypothesis, a multilayer composite OC graft was fabricated using gelatin and oxidized alginate (OXA) compositions with and without HAP for the bone and cartilage regions, respectively, and a gradient of HAP was formed in between. The two layers were formed using a 3D bioprinting method, while a porous electrospun mesh of polycaprolactone was placed in the graded region between cartilage and bone to represent the tidemark. The change in mineral content across the rabbit OC interface tissue and the OC graft interface was investigated using energy dispersive X-ray (EDX) and micro computed tomography (CT) characterization. The printability of the bioinks was verified by a strain sweep test, and volumetric expansion of both inks, with and without HAP, was examined using a swelling test. Findings revealed that both bioinks exhibited a shear thinning behavior. In addition, swelling test showed that both inks possessed similar volumetric expansion when immersed in water, demonstrating its feasibility to be used as a defect filler. EDX scan for calcium (Ca) and phosphorus (P) verified the gradient of mineral in both OC grafts and native rabbit OC tissue. The CT characterization verified a HAP gradient created in the OC graft within 168m thickness similar to the mineral gradient thickness determined for rabbit OC interface. Furthermore, the electrospun membrane was found to have pore diameters less than 1m that is sufficient to prevent vascular invasion of the articular cartilage tissue. Overall, the OC graft fabricated using combined bioprinting and electrospinning techniques demonstrated a potential to serve as a biomimetic hydrogel filler for regenerating OC defects to restore the function of the knee joint. It is expected that the proposed OC graft will be effectively used to address a significant clinical problem that affects millions of people, with significant societal and economic impacts.
Restricted
60 GHZ PHASED ARRAY PHASE SHIFTER DESIGN FOR 5G APPLICATION
(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-24) Shaimerden, Yernur
The purpose of this research work is to design a phase shifter and antenna for integration into a 60 GHz phased array transceiver designed specifically for 5G applications. A phased array transceiver consists of an active device (e.g. amplifier), phased array matrix, and antenna. This thesis focuses on the design of a 60 GHz phased antenna array with no active device. A butler matrix is implemented for the phased array matrix. In the methodology part, the development process of the butler matrix and patch antenna is presented. In this study, various types of beamforming networks are examined and written in the literature review section. The fundamental components of the Butler matrix are systematically designed, showing each step of the process. At the implementation step, CST software was used. Design is implemented on a substrate material known as Rogers RT/duroid 5880, with a 0.127 mm thickness. The results indicate the good reflection coefficient at the operating frequency of 60 GHz. Proposed design with patch antenna results in four orthogonal beams, each directed at +5°, +37°, -37°, and −5°.
Restricted
ACTIVE OBJECT TRACKING USING REINFORCEMENT LEARNING
(Nazarbayev University School of Engineering and Digital Sciences, 2022-05) Alimzhanov, Bexultan
The concept of "smart cities" has rapidly emerged as the means by which urban planners can improve the quality of life of citizens, providing better services at lower cost. Typical objectives include the optimization of traffic routing, the automatic detection of emergency "events" and related improvement in the response time of emergency services, and overall optimization of resource allocation and energy consumption. A core component of the smart city concept is the widespread deployment of closedcircuit cameras for purposes of monitoring and event detection. A typical application is to locate and track a vehicle as it moves through crowded urban scenarios. Usually, tracking and camera control tasks are separated, which induces problems for the construction of a coherent system. Reinforcement learning can be used to unify the systems, such that control and tracking can be resolved simultaneously. However, there are issues related to the collection and use of comprehensive real-world data sets for purposes of research. To avoid this problem, it is feasible to conduct the agent training using synthetic data, and then transfer the results to real-world settings. This approach also serves to address the issue of domain invariance. For the thesis, I investigate active object tracking using reinforcement learning by first developing a synthetic environment based on the videogame Cities: Skylines, using the extensive Unity engine, which accurately simulates vehicle traffic in urban settings. The complete system consisting of a trained object detector and a reinforcement learning agent is tuned in this environment with corresponding reward functions and action space. The resulting agent is capable of tracking the objects in the scene without relying on domain-specific data, such as spatial information. The thesis includes the creation of the synthetic environment, the development of the agent, and the evaluation of the resulting system.
Open Access
ADDRESSES STANDARDIZATION AND GEOCODING USING NATURAL LANGUAGE PROCESSING
(Nazarbayev University School of Engineering and Digital Sciences, 2022-07) Mussylmanbay, Meiirgali
Geocoding, the process of converting the textual addresses into a pair of coordinates, is a preliminary step in spatial analysis. However, converting addresses into latitude and longitude is not a trivial task as they are presented as arbitrary text, mostly lacking completeness, and do not follow a concrete fixed structure. Therefore, the thesis discusses the theoretical fundamentals of textual data normalization and standardization techniques and presents adequate practical approaches to how addresses written in various ways can be brought to a single standard. For binding the textual addresses with their appropriate geocodes, we have conducted practical experiments using the data collected from 5 publicly available sources and such tools as Elasticsearch, including its built-in BM25 similarity algorithm, as well as a state-of-the-art algorithm - BERT. Also, we have admitted Open Street Map address structure as a golden standard and cosine similarity algorithm as a text similarity algorithm. The practical outcomes of the models were verified on randomly chosen 100 records. The results were visualized on the map to illustrate the applicable cases of geocoding usage. Further, the raw address data and address standardization results serve as train and test data to predict the closest address and adequate geocodes for given arbitrary address representations. For the thesis, we used models based on Transformer architecture, namely T5 and BART, for predicting ’correct’ addresses. In addition, BLEU was used as a reference metric to compare the models’ accuracy. Overall, the thesis can boast rich theoretical background information and be a practical reference to how clean addresses can be revealed using state-of-the-art models given non-standard addresses.
Open Access
Adiabatically Tapered Fiber-Optic Microsensor: Fabrication and Characterization
(Nazarbayev University School of Engineering and Digital Sciences, 2020-05) Yelikbayev, Sultan
As of late, the various techniques from the materials science and biophysics are utilized to study the physical properties of the microstructure of the chemical and biological specimen. Through a considerable many of them give phenomenal sensibility, they have a few requirements concerning electromagnetic interference, fabrication complexity and specific laboratory conditions for operating. Hence, to overcome these constraints the new family of micro-scale fiber optical sensors was introduced. There are several methods to fabricate microfibers with different microtechbology. However, the techniques with more simple, economical and robust fabrication process are still developing. Therefore, this study proposes the fabrication process of widely interested tapered fiber optic microsensor via Laser Splicing System and characterization of produced tapered microfibers in terms of external RI sensitivity. The final product that was achieved is the fiber with the least waist diameter 19 mm has RI sensitivity of 156.8215 nm/RIU and agreeable linear correlation between the wavelength shift and RI change. In addition, the detailed fabrication process with characterization method is presented in the following sections. Moreover, the observed trends in fabrication process and recommendations based on the practice experience are also suggested in this report.
Restricted
ADOPTING VISUAL ODOMETRY FOR INDOOR MOBILE ROBOT LOCALIZATION
(Nazarbayev University School of Engineering and Digital Sciences, 2022-04) Sharipov, Madiyar
Localization problem is one of the primary problems for robot indoor navigation. This work discusses different types of odometry and focuses on the visual one. It implements the localization system based on the RTAB-map and compares it with such a classical approach as AMCL. The paper contain the detailed explanation of robot design for localization system implementation.
Restricted
Advanced Circuit Configurations for RF Wireless Power Transfer
(Nazarbayev University School of Engineering and Digital Sciences, 2024) Azhmuratov, Serik
Open Access
ADVANCED CIRCUIT CONFIGURATIONS FOR RF WIRELESS POWER TRANSFER
(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-26) Kudaibergenova, Zhanel
Technology for wireless power transfer (WPT) has gained more importance in the contemporary world. The upsurge spike can be a result of the WPT system's ability to power devices without the use of traditional connections. In particular, near-field WPTs have a wide range of applications, including wireless sensors, IoT, biomedical implants, RFID, and consumer electronics. It is essential to emphasize that the WPT system can be realized in a number of ways, one of which is a defected ground structure technique. This approach is well-established for its simple design process and compact system. Despite this recently developed DGS-based WPTs demonstrate poor performance, in other words, low power transfer efficiency in practical validations. The inevitable factors, such as imperfections of lumped elements, the in-house fabrication, and energy losses during transfer, have an impact on the experimental results. Therefore, various performance enhancement strategies have to be considered to realize the compact and efficient WPT system. In this regard, one of the promising methods for improving WPT operation is the use of metamaterial, which is an artificial material with unique electromagnetic features. As a result, this thesis work focuses on the development of compact and efficient WPTs applicable to various fields and on performance enhancement strategies based on metamaterial utilization
Open Access
Aerosol formation in CO2 capture plants - molecular dynamics simulation
(Nazarbayev University School of Engineering and Digital Sciences, 2017-12) Mansurov, Ulan
Carbon dioxide capture is becoming a major concern not only from the perspective of traditional sour gas sweetening but also because of adverse effects of CO2 on climate change. The most conventional method to eliminate CO2 is carried out in a post-combustion CO2 capture (PCCC) column using aqueous monoethanolamine (MEA) as a solvent. Numerous reports have manifested significant amount of solvent losses due to formation of aerosols in PCCC columns. This research provides insights into formation mechanisms of aerosols or particulate matter (PM) at a molecular level by emphasizing interaction parameters between participating components. Molecular dynamics (MD) simulations were performed using GROMACS software. Five different systems under ordinary PCCC conditions were considered each of which has unique configuration of components. MD simulations revealed evolution and development of molecular clusters that formed PM which consisted of all gaseous MEA, SO2, major portion of CO2, and water vapor. Furthermore, quantitative analysis of the molecular clusters was carried out in terms of CO2 molecules. Nucleation rates of PM were in the order of 10-30 cm-3s-1. Also, formed aerosol particles were structurally examined using radial distribution functions (RDF) and determining pair potentials between the molecules. It was found that MEA in vapor phase contributes to PM formation. Furthermore, strong attraction potential between water and CO2 and MEA imply that the presence of water in vapor phase might be one of the key factors that forms and sustains PM. Taken together, the results are first of the efforts to understand PM (aerosol) formation in a typical PCCC column based on molecular simulations, and based on the findings of the study, certain practical suggestions were offered to avoid formation of PM.
Open Access
Aerosol formation in CO2 capture plants – aspen plus simulation model
(Nazarbayev University School of Engineering and Digital Sciences, 2017-12) Galymzhanov, Nursultan
One of the most promising technologies available for decreasing CO2 concentration in the atmosphere is Post Combustion CO2 Capture (PCCC). The process is based on absorption-desorption of carbon dioxide by a solvent. Amine based aqueous solutions are considered as the state of the art solvent for PCCC. However, its use is associated with MEA emissions from an absorber column through vapour and aerosol phases. Aerosol emission has only recently been detected, and reported to be related to the degree of supersaturation of gas. The objective of this study was to develop a new conceptual model to estimate heat and mass transfer rates between gas and particulate phases using Aspen Plus simulation software. Also, validation of the model was performed by comparing it with results of an experimental mini-plant developed by TNO group in Netherlands. In the model presented in this study, interaction between the gas and the solvent, and the gas and the particles was split by modelling the gas-solvent interaction in the absorber and the gas-particles interaction in separate absorber columns representing sections of a discretised absorber. A method was presented to estimate particle formation due to nucleation and to correct the MEA loss predicted by Aspen Plus. The CO2 removal efficiency was estimated to be 95%. The estimated total molecular mass transfer rate from the gas phase at the top of the absorber column to the particle phase was found to be -7.3×10-10 kg/s, indicating net molecular mass transfer from the particle to the gas phase. The mass transfer due to nucleation was estimated to be 1.92487×10-6 kg/s. The amount of particle phase MEA emission was found to depend on the temperature inside the absorber, temperature bulge, gas supersaturation ratio, volume of particles entering the absorber and H2SO4 concentration in the entering gas. The particle phase MEA emission due to the molecular mass transfer from the gas phase to the particle phase was found to be 0.3 mg/Nm3gas, while particle phase MEA emission resulted from the nucleation mass transfer was 697.0 mg/Nm3gas. Thus, the total particle MEA emission was estimated to be 697.3 mg/Nm3gas. The estimated nucleation rate is approximately 2×1015 particles.cm-3.s-1. Gas phase MEA emission was found to be 1.3 mg/Nm3gas.
Restricted
Aerosol formation in ethanolamines based post combustion CO2 capture plants
(Nazarbayev University School of Engineering and Digital Sciences, 2019-01) Saparov, Ablay; Torkamahalleh, Mehdi Amouei; Shah, Dhawal
In recent years, capture of carbon dioxide released from fuel-combustion has become one of the most important field operations. A commonly used process for carbon dioxide capture is applying a post-combustion-CO2-capture (PCCC) column with solvents like aqueous methyldiethanolamine (MDEA), monoethanolamine (MEA), and other amines. There are several reports indicating formation of aerosol in amine based PCCC columns, which in turn is associated with solvent losses. The purpose of this work was to investigate the mechanism of nucleation (leading to an aerosol or particulate matter formation) for three solvents (MEA, MDEA, and a mixture of MEA and MDEA) using the molecular dynamic simulations. Additionally, two cases with different solvent compositions, corresponding to industrial usage, were simulated. Using the simulations, role of different solvents were analyzed in order to identify the mechanism and rate of aerosol formation. The results provide insights to the structure and composition of the aerosol. In particular, we observed a highest growth for the case with aqueous MDEA solution, whereas nucleation was fast for MEA solutions. Furthermore, the rate of formation of aerosol increased with increasing CO2 concentrations. The formed aerosols mainly consisted of water, CO2, and the ethanolamines. Taken together, the result from this work contributes to better understand aerosol formation in PCCC columns.
Restricted
AIR-TO-GROUND CHANNEL MODELING FOR UNMANNED AERIAL VEHICLES
(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-19) Mukhatzhanova, Sabina
The ‘Air-to-Ground Channel Modeling for Unmanned Aerial Vehicles’ main goal is to develop a precise and reliable model of the communication channel between unmanned aerial vehicles and ground stations. This project will be mainly focused on the extensive research of recent data and channel simulation using MATLAB. The project will analyze different scenarios and characteristics that may have an impact on the proper work of the channel and find ways to overcome those difficulties. These obstacles can be channel fading, polarization, weather conditions, and hard-to-reach locations. The proposed channel modeling can be beneficial in different spheres of life: safety, surveillance, telecommunication, and national priorities. Overall, the project’s main focus will be based on the enhancement of analytical skills in order to select major types of channel modeling and see the advantages and disadvantages of the proposed model. The results can have a positive impact through the contribution to the advancement of UAV technologies.
Restricted
ANALYSIS AND COMPARISON OF APPROXIMATE K-NEAREST NEIGHBOR ALGORITHMS
(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-19) Aidarbek, Angsar
This study presents a comprehensive comparison of Approximate k-Nearest Neighbor (AKNN) methods across multiple datasets, including image, text, and behavioral datasets. The performance of various AKNN algorithms is evaluated in terms of build time, search time, total time, and recall metrics for different datasets. Key findings reveal that tree-based AKNN methods exhibit vulnerability to changes in dataset contents, while graph-based algorithms demonstrate superior performance in certain scenarios. Furthermore, algorithm-specific nuances, such as computational efficiency and recall rates, are discussed across diverse datasets. Insights from this study provide valuable guidance for selecting suitable AKNN methods based on specific application requirements and dataset characteristics. Furthermore, potential directions for future research, including scalability improvements, algorithmic enhancements, and domain- specific applications are identified to further advance the field of AKNN algorithms.
Embargo
ANALYSIS AND OPTIMIZATION OF TRANSPORT AND REACTION PROCESSES IN THE CYLINDRICAL FLOW-THROUGH CATALYTIC MEMBRANE REACTORS
(Nazarbayev University Graduate School of Engineering and Digital Sciences, 2024-04-22) Abbas, Qaiser
Cylindrical flow-through catalytic membrane reactors, employing porous membranes impregnated with catalysts, offer enhanced selectivity and yield in chemical reactions. This work is focused on mathematical modeling and numerical analysis of a cylindrical flow-through catalytic membrane reactor. The reactor's geometry, which incorporates a porous membrane, is specially tailored for the cylindrical configuration, addressing a research gap in realistic geometries. The proposed mathematical model includes a series of irreversible reactions with power-law kinetics occurring under non-isothermal conditions. A system of non-linear diffusion-convection-reaction equations is formulated for a cylindrical catalytic membrane reactor under both steady and unsteady-state. The study investigates the occurrence of dead zones within the membrane reactor as a result of rapid reactant depletion, a phenomenon that has not been extensively studied in prior literature for cylindrical membrane reactors. Problems with fractional reaction exponents require efficient numerical solvers since conventional iterative solvers encounter difficulties due to the fact that the power-law reaction term with fractional reaction exponent is not differentiable at the vanishing concentration. A novel time-marching scheme specifically designed for the cylindrical catalytic flow-through membrane reactor is developed and applied for simulations to get valuable insights into dead-core phenomena. The effects of dimensionless process parameters such as Thiele modulus, mass Peclet number, heat Peclet number, etc. and a model parameter (i.e., geometry parameter) on the concentration and temperature profiles, as well as dead-zone formation, are extensively investigated under steady-state. The simulation results demonstrate that these parameters affect the occurrence of dead zones and their size. The impact of convective flow on the reactor performance indicators under steady-state is also presented. Moreover, the investigation extends to unsteady-state conditions, exploring the dynamic behavior of concentration and temperature profiles as well as productivity under both isothermal and non-isothermal scenarios. In the case of a single reaction, the analysis of productivity reveals substantial percent increments, offering insights into the identification of optimal conditions. Finally, a comprehensive exploration into the optimal conditions for productivity in a sequential reaction is conducted.
Open Access
ANALYSIS OF COVID-19 DATA AND PREDICTING FUTURE CORONAVIRUS CASES BY USING MACHINE LEARNING ALGORITHMS
(School of Engineering and Digital Sciences, 2023) Zhaksybay, Uldana
Background: The Covid-19 pandemic has posed significant challenges to healthcare systems worldwide. Effective strategies to manage the pandemic require accurate and timely forecasting of the spread of the virus. Machine learning (ML) algorithms offer a promising approach for predicting the number of Covid-19 cases. Objectives: This thesis work aims to analyze the Coronavirus data, and the number of cases and predict the future behavior of Covid-19 in Kazakhstan which helps to make key decisions related to the virus and prevent the country from the global economic crisis. Methods: The study utilized publicly available data sources to create a comprehensive Covid-19 dataset. The dataset included daily counts of confirmed Covid-19 cases, deaths, recoveries, and tests across multiple countries and regions worldwide. This work used four ML algorithms in our study, including a decision tree, random forest, linear regression (LR), and polynomial regression. Evaluation of the performance of the models based on r2 score, MAE, MSE. Results: Results showed that all four ML algorithms produced reasonably accurate predictions of Covid-19 cases. The random forest and decision tree algorithms outperformed the other models, with an accuracy rate of over 85% and 90% respectively. The linear and polynomial regression models had accuracy rates of approximately over 75%. Conclusion: In conclusion, this study demonstrates the potential of ML algorithms for predicting the number of Covid-19 cases. Findings suggest that the random forest algorithm is the most effective in forecasting Covid-19 cases. The results of this study may help inform policymakers and healthcare professionals in developing effective strategies to manage the Covid-19 pandemic.
Embargo
ANALYSIS OF EFFICIENT REGIONS FOR WIND POWER GENERATION IN KAZAKHSTAN
(Nazarbayev University School of Engineering and Digital Sciences, 2024-05) Aryngazin, Anuar
Global concerns about climate change and the decreasing of fossil fuels reserves have made renewable energy production a major priority. Wind energy is a well-established research area. However, identifying optimal regions for wind power production requires weighing multiple factors, including wind speed, density, availability, and the environmental consequence of reducing greenhouse gas (GHG) productions, and application of decision analysis. The main purpose of this study is to analyze the regions of Kazakhstan and identify the optimal location for wind energy production using the decision analysis extension of multiattribute utility theory (MAUT). Previous literature conducted analyzing wind power potential in Kazakhstan has identified several promising locations for efficient wind energy production. However, the selection process did not involve mathematical analysis; instead, the choice was primarily determined by quantitative results obtained from empirical studies. This study identified seven favorable areas based on expected utility and a comprehensive study was conducted using the multiattribute utility function (MUF). This method facilitated the selection of wind energy projects in efficient regions, taking into account factors such as potential power production and GHG emission reduction potential. The study involves collecting data on the GHG emission reduction potential and wind speed, followed by analysis and decision making to identify practical wind energy projects. Moreover, the data analysis includes a questionnaire for the decision maker (DM). This made it possible to determine the optimal locations for installing wind farms based on the real preferences of an expert in the field of energy. The results based on the expected utility values showed that the most optimal location for installing a wind power plant is Fort-Shevchenko and Yereymentau. The results of the research project are critical for the development of wind energy in Kazakhstan and for identifying the most efficient regions for wind energy production using utility theory. This study contributes to the development of wind energy in Kazakhstan and provides information for decision-making processes on sustainable energy production.