01. PhD Thesis
Permanent URI for this collection
Browse
Recent Submissions
Item Open Access TIO2- AND AG2CO3-BASED PHOTOCATALYSIS FOR REMOVING EMERGING POLLUTANTS FROM WATER(Nazarbayev University School of Engineering and Digital Sciences, 2024-09-04) Mergenbayeva, SauleWater plays a vital role as an essential natural resource that supports the development of life and human endeavors. Large volumes of water are consistently polluted with various pollutants, including emerging pollutants (EPs). EPs are a group of persistent pollutants like pharmaceuticals and personal care products (PPCPs), dyes, and endocrine-disrupting compounds (EDCs). Many of them result from overuse, and their presence can cause serious health issues. However, conventional water treatment technologies are not specifically designed to remove these kinds of pollutants. In this study, advanced oxidation processes (AOPs) utilizing newly developed TiO2 and Ag2CO3-based photocatalysts were applied to remove several model EPs. The catalysts were synthesized using various preparation methods and then characterized to investigate their crystal structure, morphology, elemental composition, and optical characteristics using XRD, Raman spectroscopy, SEM, TEM, EDS, BET analysis, and UV-Vis DRS spectroscopy. Firstly, TiO2 combined with Ti2O3 (mTiO), calcined at various temperatures, was used to degrade 4-tert-butylphenol (4-tert-BP) under simulated solar light. Among the catalysts, mTiO calcined at 650°C (mTiO-650) achieved the highest degradation and mineralization performance, achieving approximately 90% and 50%, respectively. The phase transformation and surface area of the calcined catalysts played a significant role in enhancing their photocatalytic activity. Additionally, the presence of various anions (NO3−, Cl−, HCO3− and CO32−) and humic acid (HA) in water was evaluated for its effect on 4-tertBP degradation. In addition, potential pathways for the degradation of 4-tert-BP were suggested. This study marks a significant advancement with the successful preparation of mTiO-650 catalyst capable of eliminating of 4-tert-BP in water. Secondly, mono- and co-doped TiO2 catalysts were synthesized using simple impregnation methods. Mono-doped Mo@TiO2 and W@TiO2 improved the adsorption capacity and reduced the energy gap (Eg), while co-doped catalysts, namely CuMo@TiO2 exhibited distinct adsorption properties and slightly enhanced degradation performance. CuMo@TiO2 fully degraded 4-tert-BP under UV light in 60 min and achieved 70% degradation under solar light in 150 min, marking the first application of both mono-doped and co-doped TiO2 catalysts for degrading 4-tert-BP in water, emphasizing their potential in environmental remediation, and introducing a novel pathway for efficient photocatalyst synthesis. Moreover, TiO2 doped with Fe (Fe@TiO2) was successfully synthesized via the wet impregnation method. Surface morphology was analyzed using SEM and TEM, while EDS confirmed the incorporation of Fe into TiO2. Fe@TiO2 demonstrated enhanced light absorption in the 200–365 nm range compared to bare TiO2. The photocatalytic performance of the catalyst was assessed in a continuous flow system for the degradation and mineralization of 4-tert-BP under UV light (254 nm). Fe doping slightly improved the degradation efficiency, with 87% of 4-tert-BP degraded in 60 min, compared to 82% with bare TiO2. Furthermore, Fe@TiO2 catalyst achieved 49.6% mineralization of 4-tert-BP. Additionally, Ag@TiO2 catalysts were also developed for reducing 4-nitrophenol (4-NP) to 4-aminophenol (4-AP), with Ag@TiO2-21 achieving 98.3% 4-NP reduction within 10 min. Moreover, Ag@TiO2-21 displayed strong antibacterial activity, with minimum inhibitory concentration (MIC) values ranging from 0.5 to 1 mg/mL against B. subtilis, E. coli, S. aureus, and P. aeruginosa. This study introduces Ag@TiO2 catalysts for efficient elimination of 4-NP from water, showcasing significant antibacterial activity. This research highlights the potential of Ag@TiO2 catalysts for environmental applications and offers a novel approach to developing effective photocatalysts. Furthermore, TiO2@zeolite (TiO2@Z and TiO2@ZSM) catalysts were tested for the degradation and mineralization of sulfamethoxazole (SMX) under UV (365 nm) light. The TiO2@ZSM1450 catalyst achieved complete SMX degradation within 10 min and mineralized 67% of SMX in 120 min. The effect of anions (NO3−, Cl−, and CO32−) on the photocatalytic performance of the TiO2@ZSM1450 catalyst was also investigated. The study highlights the effectiveness of modifying TiO2 with zeolite for improved photocatalytic performance. Lastly, Ag2CO3 was employed for the photocatalytic degradation of 4-tert-BP, achieving 100% degradation within 60 min under solar light. The effect of key parameters, including 4-tert-BP concentration (2.5–10 mg/L), Ag2CO3 dosage (100–300 mg/L), various light sources, and the presence of different anions, was studied. The re-usability of Ag2CO3 remained effective after three successive experimental runs. This study introduces Ag2CO3 as a novel catalyst for 4-tert-BP elimination in water, emphasizing its potential in environmental applications and offering a new synthesis approach. The results of this Thesis have great significance in advancing water treatment technologies and provide original and novel alternatives for the elimination of EPs in water.Item Open Access MODEL PREDICTIVE CONTROL AND IMITATION LEARNING ALGORITHMS FOR ROBOT MOTION PLANNING IN PHYSICAL HUMAN-ROBOT INTERACTION(Nazarbayev University School of Engineering and Digital Sciences, 2024-08-07) Aigerim NurbayevaThis PhD thesis focuses on the design and testing of safe robot motion planning algorithms for human-robot workspace sharing. These algorithms are based on the use of nonlinear model predictive control (NMPC), a model-based method for motion planning relying on numerical optimization. The contribution of the thesis can be split into two main areas. The first area consists of the approximation of NMPC laws using deep neural networks (DNNs), often referred to as “imitation learning”. This is motivated by the fact that the execution of NMPC laws might require a considerable amount of time, which restricts the performance of the closed-loop system. Calculating the output of a DNN for a given input is instead a much faster process. Therefore, replacing the optimization solver of NMPC with a DNN can reduce computation times, thus improving performance. It is crucial, though, to suitably train the DNN to imitate the NMPC law in order to improve performance and at the same time guarantee safety. The final result obtained in this area consists of using the so-called dataset-aggregation approach for DNN training, together with properly designed safety filters, which ensure that the safety constraints imposed in the NMPC problem also hold for the robot motion generated by the DNN. The second area consists of the extension of a previously defined NMPC law in terms of stabilizing terminal constraints. The most common approach for guaranteeing closed-loop stability in an NMPC problem is the imposition of terminal constraints, i.e., the prediction of the system motion is required to satisfy certain conditions at the end of the prediction horizon. Specifically, in a previous approach, the “point terminal constraint” was used, in which the prediction of the robot motion had to exactly reach the desired goal configuration at the end of the prediction horizon. In this thesis, this condition is relaxed by imposing that a given set, rather than a given point, is reached in the state space for the predicted robot motion. The imposition of this new condition allows for an enlargement of the domain of attraction, i.e., the NMPC law can find a solution for reaching the goal configuration from a wider set of initial configurations. All the proposed motion planning strategies were tested experimentally on a UR5 collaborative manipulator.Item Open Access MULTISCALE MODELING OF CHEMICAL STABILITY AND TRANSPORTATION OF OH- ION FOR CHITOSAN-BASED BIOCOMPOSITE ANION EXCHANGE MEMBRANE FUEL CELLS(Nazarbayev University School of Engineering and Digital Sciences, 2024-08) Karibayev, MiratAnion Exchange Membrane Fuel Cells are obtaining popularity in current research due to their promising advancements, which include low production costs, the ability to use catalysts free of platinum group metals, moderate operating temperatures, and high power densities. However, the primary challenge of Anion Exchange Membranes is associated with chemical instability of the quaternary ammonium head groups in alkaline conditions and elevated temperatures, which also led to a decrease in the diffusion of hydroxide ions. This study used Density Functional Theory calculations, ab initio Molecular Dynamics simulations, and conventional all-atom Molecular Dynamics simulations to examine the chemical stability of different chemical structures, including quaternary ammonium head groups, quaternized chitosan head groups, and Deep Eutectic Solvent supported quaternized chitosan head groups as well as the diffusion of hydroxide ions. This research work consisted of the following four main objectives: i) the degradation mechanisms of different quaternary ammonium head groups under different hydration levels via the Density Functional Theory method, ii) the diffusion of hydroxide ion via different quaternary ammonium head groups under different hydration levels via conventional all-atom Molecular Dynamics simulations, iii) the degrataion mechanisms of various quaternized chitosan head groups and the diffusion of hydroxide ions under different hydration levels and temperatures via the Density Functional Theory method and conventional all-atom Molecular Dynamics simulations, and finally iv) explore the degradation mechanisms and diffusion mechanisms of hydroxide ion via Deep Eutectic Solvents supported tetramethylammonium head group and quaternized chitosan head group under different hydration levels and temperatures via Density Functional Theory calculations and ab initio Molecular Dynamics simulations....Item Restricted EVAPORATION AND WATER BALANCE OF SMALL ENDORHEIC LAKES IN SEMI-ARID NORTHERN KAZAKHSTAN(Nazarbayev University School of Engineering and Digital Sciences, 2019) Yapiyev, VadimApproximately two thirds of global precipitation falling over continental surfaces is reverted to the atmosphere by terrestrial evaporation. Over the terrestrial surfaces, the difference between Precipitation-Evaporation (P-E) is stored as soil- surface- and groundwater and contributes to surface and sub-surface runoff that ultimately returns water back to the ocean by stream and groundwater flow. Chapter 1 sketches the global water cycle and underlines a relative importance of evaporation in endorheic basins. Endorheic basins (i.e., land-locked drainage networks) and their lakes can be highly sensitive to variations in climate and adverse anthropogenic activities, such as overexploitation of water resources. Chapter 2 provides a brief overview of one major endorheic basin on each continent, plus a number of endorheic basins in Central Asia (CA), a region where a large proportion of the land area is within this type of basin. In CA a substantial increase in irrigated agriculture coupled with negative climate change impacts have disrupted the fragile water balance for many endorheic basins and their lakes. Transboundary integrated land and water management approaches must be developed to facilitate adequate climate change adaptation and possible mitigation of the adverse anthropogenic influence on endorheic basins. Subsequently, the focus shifts to the endorheic lakes within Burabay National Nature Park (BNNP), Northern Kazakhstan (the main focus of this thesis). These endorheic lakes have been drying out during the last one hundred years or so with a public perception that the water level decrease accelerated in the past few decades.Item Open Access IMPEDANCE BASED APTASENSOR FOR THE DETECTION OF MYCOBACTERIUM TUBERCULOSIS SECRETED PROTEIN MPT64(Nazarbayev University School of Engineering and Digital Sciences, 2019) Sypabekova, MarzhanTuberculosis (TB) detection remains a significant healthcare issue in the developing world owing to a number of challenges. Current diagnostics are based on microbiological culturing, sputum smear microscopy, and nucleic acid amplification tests. These methods suffer from limitations such as batch to batch variations, frequent contaminations, low sensitivity, and the requirement for special facilities, expensive devises, reagents, and trained personnel. This thesis describes the development of the sensitive oligonucleotide-based aptasensor for the detection of TB biomarker MPT64 protein. The dissertation investigates the selection and use of ssDNA aptamers to detect MPT64 using the electrochemical impedance spectroscopy (EIS). Aptamers serve as bio-recognition elements in this study, and they have numerous advantages including cheap cost, ease of modification and long shelf life. The combination of aptamers with the EIS offers sensitive detection since the change in EIS signal can be recorded as the result of analyte binding event based not only on molecular interaction level but also on electron transfer levels. As the result 17 unique aptamer sequences were purified and analyzed. One aptamer with dissociation equilibrium constant KD of 8.92 nM was selected and the surface chemistry was optimized based on ssDNA aptamer modified with a long linker and 6-mercaptohexanol as a co-adsorbent at 1/100 ratio. The selected aptamer was further immobilized on an interdigitated microelectrode and connected to a portable potentiostat. The detection time for aptasensor was found to be 15 min. The aptasensor was tested on clinical samples and showed increased binding to TB (+) samples as compared to TB (-) samples. The integration of the aptasensor with the in house built fluidic chamber and liquid flow rate within chamber was also investigated. The work in this thesis is significant as it can contribute to the diagnosis of TB (non-invasive), monitoring of anti-TB treatment in infected people and hence to socio-economic development of the country. It is the first portable aptasensor which is developed using aptamers and EIS as a detection technique that can provide fast clinical sample analysis (reduced from 3 h to 15 min) as well as elimination of using of extra reagents, equipment, and personnel.Item Open Access OPTIMAL DESIGN AND CONTROL OF VARIABLE IMPEDANCE ACTUATED ROBOTS(Nazarbayev University School of Engineering and Digital Sciences, 2019) Zhakatayev, AltayIn this thesis, the challenging problems of design and control of variable impedance actu ated robots are considered. The difficulties arise due to nonlinear dynamics, physical con straints of the system, and presence of additional actuators and nonlinear elastic/damping elements. As a result, we propose a control methodology, which takes into account system constraints and input bounds, guarantees system utilization to its full potential, and closely achieves the system’s target performance level. The thesis consists of seven chapters. The first chapter gives a broad introduction to the problem and provides the literature review. For example, differences between position-controlled robots and variable impedance actu ated robots are discussed, their corresponding advantages and disadvantages are presented and compared, past design and control solutions are reviewed, and the hypothesis is de scribed. The second chapter covers the proposed closed-loop control methodology for variable stiffness actuated robots. This chapter covers the general idea behind closed-loop control of variable impedance actuated robots using model predictive control, and it also includes simulations and experimental results. The augmentation of the variable stiffness robots with reaction wheels is described in chapter three. Specifically, the advantages of using reaction wheels to actuate the variable stiffness robots are discussed. This is fol lowed by a discussion of time-optimal control of variables stiffness robots in chapter four. This chapter presents and describes two time-optimal control problems: minimum time for target performance and minimum time for maximum performance. In chapter five energy optimal control of variable stiffness robots is described. In particular, three energy-optimal control problems are defined: maximum performance with limited energy, target perfor mance with minimum energy and maximum performance with minimum energy. Then chapter six contains successive linearization-based model predictive control of variable stiffness robots. The main idea of this chapter is that linearization might be beneficial for model predictive control of nonlinear systems due to a simpler model and the resulting smaller sampling time. Finally, chapter seven describes the potential impact of our research in the field of robotics and society.Item Open Access ELECTROANALYSIS OF MICROBIAL BIOFILMS AND ANTIBIOFILM DRUG TESTING(Nazarbayev University School of Engineering and Digital Sciences, 2022) Olaifa, Kayode WilliamMicrobial biofilms are responsible for about 80% of infectious diseases in humans, resulting in high morbidity and mortality rates. Biofilm confers protection to the microbial cells from stressors, including antimicrobial treatments. Biofilms are more difficult to remove/kill, thus, contributing to antimicrobial resistance phenomenon. Unfortunately, existing methodologies routinely employed for microbial biofilms and evaluation of anti-biofilm compounds are flawed with varying limitations. Therefore, an urgent need for the design/development and adoption of new diagnostic platforms is exigent. Additionally, the need for new therapeutic options cannot be overemphasized. In this work, a bioelectrochemical platform that uses simple, low-cost, and commercially available screen-printed electrodes was implemented for real-time evaluation of selected antimicrobials against clinically relevant biofilm-forming species. We also adopted a drug repurposing strategy against a model resistant bacterial strain using the developed bioelectrochemical platform. Finally, attempt was made to detect a model fungal pathogen in human urine samples also via the developed platform. In general, both biochemical and electroanalytic methods suggests that complete inhibition of biofilm formation would require concentrations higher than that needed for planktonic cells. Further optimization of the methodology on C. albicans biofilms indicated that the antifungal activity of the tested compounds is in the order of complex Ag3>Amphotericin B>Fluconazole, while the conventional XTT indicated the order of Amphotericin B > Fluconazole >complex Ag3. This variability further reiterates the necessity for a multi-method approach to validate the antibiofilm efficacy of any compound. However, this study demonstrated, for the first time, the real-time antibiofilm assessment of selected antimicrobials using electroanalytical approach and offers consistent findings as early as 10 h following inoculationItem Embargo EXPERIMENTAL INVESTIGATION OF NANOPARTICLES FOR CANCER CARE AND APPLICATION IN BIOSENSING AND THERMAL ABLATION(Nazarbayev University School of Engineering and Digital Sciences, 2022-07) Ashikbayeva, Zhannatis related to the class of disorders characterized by abnormal cell proliferation that can spread and damage other human organs. There are diverse forms of cancer based on the position of a tumor in a human body. The most prevalent cancer types diagnosed in patients are breast, liver, lung, and prostate with high mortality rates worldwide. Surgery, chemotherapy, and thermal ablation (TA) therapy are known as conventional treatment techniques for the treatment of cancer. TA therapy is gaining interest in cancer cure due to its minimal invasiveness and ability to reach tumors in challenging regions to access. Moreover, not all patients surgical candidates because of medical and physical conditions. However, non-specific heat damage and non-accurate temperature monitoring are known as the main limitations during thermal therapy. Therefore, this thesis focused on the application of metallic nanoparticles during thermal procedures making it possible to rise the heat in the targeted region. Moreover, optical fibers may be employed as a sensing system temperature monitoring accurately in real time. In this work, the impact of metallic nanoparticles and metallic thin films is proposed, in order to improve cancer therapeutical and diagnostic tasks. On one side, the application of nanoparticles to cancer thermotherapies is discussed and quantitatively evaluated, in methods that employ radiofrequency, microwave, or optical power delivery methods. The heat increase during therapy was achieved and validated by several types of metallic nanoparticles, particularly iron oxide magnetic nanoparticles (IONPs), gold nanoparticles AuNPs), and silver nanoparticles (AgNPs). On the other hand, biosensors can improve the early-stage cancer diagnostic by detecting biomarkers for specific types of cancer; fiber optic sensors have shown detection at ultra-low limits, compatible with scarce analytes. Biosensors empowered by thin gold films or gold nanoparticles are discussed, showing how devices fabricated in-house can turn into highly performing devices through a functionalization step.Item Open Access PERFORMANCE-BASED APPROACH TO ASSESS DURABILITY OF REACTIVE POWDER CONCRETE(Nazarbayev University School of Engineering and Digital Sciences, 2022) Bakhbergen, UmutClaiming that a relatively new type of concrete called reactive powder concrete (RPC) is reported to have superior compressive and tensile strength, and potentially improved durability because of dense microstructure, it is suggested as a material for the pile foundation system where renewable energy is to be stored in the form of compressed air inside the hollow section of the pile. Even though strength and microstructure of RPC are studied extensively by number of researchers, and some studies focus on the external sulfate attack and freezing and thawing resistance of the material, modeling of its durability properties is barely attempted. Hence, the main aims of this research are to verify the strength and properties of RPC before the exposure to damaging environments, as well as to evaluate its behavior under the effect of sulfate and freezing and thawing damages. In addition, it is suggested that the performance of RPC under external sulfate attack and freezing and thawing damage can be predicted by comprehensive models. Thus, the objective of this research is to extend the existing knowledge of the RPC performance under different exposure conditions by studying its characteristics while varying its mixture contents, and to develop simple performance-based assessment tools to evaluate characteristics of the material.Item Open Access INNOVATIVE DESIGN AND ANALYSIS OF PERFORMANCE ENHANCED WPT/SWIPT SYSTEMS FOR LOW-POWER APPLICATIONS(Nazarbayev University School of Engineering and Digital Sciences, 2022-09-30) Dautov, KassenThe wireless power transfer (WPT) technology assists users to rid of inconvenient wires and facilitates powering and charging the devices’ batteries. The WPT systems have the potential to bring a complete turnaround in a variety of applications. They have been lately employed in a number of segments such as biomedicine, consumer electronics, low-power deceives, and wireless technologies. WPT can be broadly classified as far- (radiative) and near-field (non-radiative) types. If the former is accomplished using protocols, namely, time-switching and power-splitting relaying techniques, the latter is achieved by following any of coupling-based technique (i.e., capacitive or inductive). In this context, it is worth mentioning that the concept of magnetic resonant coupling also falls under the umbrella of the inductive coupling. This enables the synchronization of the resonance of the transmitter and receiver to enhance the WPT performanceItem Embargo ENHANCING CLIMATE MAPPING METHODOLOGIES: A NOVEL PERFORMANCE-BASED FRAMEWORK FOR KAZAKHSTAN'S BUILDING CLIMATE ZONING(Nazarbayev University School of Engineering and Digital Sciences, 2024-04) Remizov, AlexeyAccurate climate zoning is crucial in the construction sector for both building thermal performance and energy efficiency, playing a vital role in achieving energy reduction targets, and facilitating early-stage design decisions to minimize energy consumption while maintaining occupant comfort. Traditional climate classification methods, which mainly rely on climatic data, often fail to consider the energy consumption of buildings, leading to a disconnect between climate classification and building energy performance. In Kazakhstan, this mismatch is evident as existing standards and climate maps do not adequately inform about potential building energy consumption in different climate zones, which leads to a high energy usage rate, emphasizing the urgent need for effective energy efficiency measures as the building stock expands. Moreover, despite currently low energy prices and sufficient energy resources in Kazakhstan, primarily non-renewable sources like coal and gas, there's a looming concern over future energy sustainability and affordability. The uncertainty arises from the significant increase in energy prices that are already being noticed. Considering the low average household income, proactive energy-efficient practices are imperative to mitigate future energy challenges.Item Restricted ROLE OF HUMAN NEI-LIKE DNA GLYCOSYLASES IN DNA REPAIR AND CANCER RESISTANCE(Nazarbayev University School of Engineering and Digital Sciences, 2024-02) Baiken, YeldarIndividuals presenting with non-resectable and rapidly metastasizing tumors, including but not limited to breast and lung cancers, are commonly managed through the synergistic application of multi-agent chemotherapeutic regimens and ionizing radiation therapy. Chemotherapy and radiation therapy cause a type of DNA damage known as complex DNA damage (CDD). This includes large DNA adducts, links between DNA strands (interstrand DNA crosslinks), and clustered lesions, which are groups of damage that include double-strand breaks (DSBs). These complex damages are more intricate in their structure compared to individual, isolated lesions. (1). Although CDDs constitute the minor proportion of total DNA damage that most anticancer agents inflict in cellular environments, its cytotoxicity is pronounced if left unrepaired. Whilst the sensitivity of cancer cells to chemotherapy and radiotherapy is usually acquired initially, but within three-to-twelve-month time, resistance to these therapies might be developed. Hence, the resistance of tumor cells to therapeutic intervention still largely confines the optimal efficacy of chemotherapy and radiotherapy for advanced cancers. The molecular mechanisms that underpin the intricate resistance in cancer are not fully elaborated. Among potential mechanisms, the activation of DNA repair pathways presents a compelling hypothesis. Importantly, the cross-resistance that tumor cells show with various DNA cross-linking agents after treatment with a single interstrand DNA crosslink-inducing agent would indicate upregulation of a definite DNA repair process on the part of the cells as an integral component of resistance to the cytotoxicity of interstrand DNA crosslinks in DNA (2). Investigating the molecular mechanisms of DNA damage recognition by repair proteins could enhance our understanding of cellular DNA damage signaling and the coordination of repair processes. In the present project, we propose to study repair of complex DNA damage in human cancer cells. The project aims to address following questions: (i) Whether there are unidentified repair activities present in cancer cells; (ii) What are the mechanisms of DNA repair coordination in human cancer cells and whether they involve the specific protein-protein interactions, multi-protein complexes and post-translational modifications; (iii) Whether the new post-replicative modification of DNA, discovered recently during our research collaboration, is involved in the removal of complex DNA damage and in the coordination of DNA repair pathways in cancer cells.Item Open Access DEVELOPMENT OF CELL-PENETRATING NANOPARTICLES FOR DRUG DELIVERY(Nazarbayev University School of Engineering and Digital Sciences, 2024-05-27) Zhaisanbayeva, BalnurRecently, there has been a growing interest in nanoparticle-related pharmaceutical and biomedical research. Anticipated outcomes of such applications include the development of in vitro and in vivo diagnostics kits, improved biocompatible materials production, and advancing drug delivery systems. In the realm of inorganic nanoparticles, silica or materials coated with silica exhibit potential for biomedical applications due to their small size, stable chemical structure, colloidal stability, and high surface reactivity. Despite the growing interest in silica nanoparticles, little is known about their toxicity resulting from the various synthesis methods; thus, recent findings often contradict each other. Moreover, most synthesis studies need more information about nanoparticle behaviour in the physiological environment, making it challenging to understand the biological effects of these nanoparticles for further clinical trials. Therefore, a newly emerging approach, safe-by-design, is starting to play a crucial role in developing nanoparticles for biomedical sciences. This dissertation explores organosilica nanoparticles synthesised from 3-mercaptopropyltrimethoxysilane (MPTS) for potential biomedical applications as a drug delivery system. The work involves extensive characterisation and toxicological evaluation of organosilica nanoparticles with thiol groups on the surface. The experiments have underscored the safety of organosilica nanoparticles through comprehensive in vitro and in vivo assessments. The further potential use of these nanoparticles was explored by covalently attaching cell-penetrating peptide (TAT) and anticancer drugs (doxorubicin). The findings of this work demonstrated that the functionalised nanoparticles changed the function of thiolated nanoparticles, and conjugated drugs continued to be effective and retain their properties.Item Restricted CYTOSKELETON DYNAMICS AND SPATIAL ORGANIZATION DURING EPITHELIAL-TO-MESENCHYMAL TRANSITION(Nazarbayev University School of Engineering and Digital Sciences, 2024-05-16) Nurmagambetova, AsselRATIONALE: Epithelial-to-mesenchymal transition (EMT) is a process that occurs during normal physiological processes (embryogenesis and organ formation) and if it is inappropriately activated it can lead to pathological processes (formation of scars, cancer metastasis, etc.). EMT is well studied at the morphological and transcriptome level. However, cytoskeleton changes during this process are less well understood. The cytoskeleton consists of microtubules, actin filaments, and intermediate filaments. In addition, there are protein complexes named focal adhesions that provide cell attachment to the extracellular matrix, and connect the actin cytoskeleton with the extracellular matrix. To describe the changes in the behavior of the cytoskeleton, namely microtubules and actin cytoskeleton during EMT is of particular interest. In addition, describing the behavior of focal adhesions during EMT is also important. AIM: The objective of this study is to describe quantitatively morphological changes that occurred in post-EMT MCF-7, A-549, and HaCaT cells, analyze microtubule dynamics, spatial organization, and its contribution to cell motility, identify changes in actin filament organization and study focal adhesion turnover. HYPOTHESIS: The dynamics of microtubules in cells undergoing EMT v might change. Cells undergoing EMT are expected to have more dynamic microtubules. Cells undergoing EMT are expected to more efficiently adhere to diverse substrates and therefore better spread. Focal contacts in cells undergoing EMT are expected to be more pronounced and dynamic than in cells not undergoing EMT. METHODS: To study changes in post-EMT cells, EMT was induced in three different cell models: MCF-7, A-549, and HaCaT. To evaluate that EMT happened, western blot and quantitative polymerase chain reaction (q-PCR) were applied to determine the level of expression of master regulators of EMT. Cell images were recorded using bright field microscopy, and analyzed using the Fiji Image J program. In analyzed cells, microtubule networks and actin filaments were visualized by immunofluorescence. To follow, describe, and measure microtubule dynamics transfection with EB-3-RFP protein was conducted. To visualize focal adhesions, two approaches were used: transduction with a talin red fluorescent protein (Talin-RFP) and transient transfection with Ptag-RFP-vinculin. Films were recorded using time-lapse fluorescent microscopy and analyzed using the Fiji Image J program. All statistical analysis was performed using GraphPad Prism (Dotmatics, USA) and a nonparametric Mann-Whitney U test or parametric t-test with Welch correction. The actin filament measurements were vi completed using Matlab scripts. CONCLUSION: This study showed morphological changes in three post- EMT cell cultures studied. All types of cells increased in size. MCF-7 and HaCaT became spread out, while A-549 became elongated. All three post-EMT cell cultures had changes in microtubule organization and dynamics. Post-EMT MCF-7 and HaCaT cells showed microtubules at a low density at cell borders, while post-EMT A-549 cells had less covered nuclei by microtubules. In all three studied models, the microtubule growth rate increased and the length of the microtubule plus end tracks became longer. The average angle of microtubule growth trajectories to cell radius decreased. Actin fibers rearranged into stress fibers in post-EMT cells. The area of focal adhesions decreased in all post-EMT cell cultures studied and focal adhesions appeared localized throughout the inner areas of spread cells. These results indicate that cytoskeletal changes make a significant contribution to the EMT process.Item Restricted EFFECTS OF SCANNING TRAJECTORY AND PARAMETERS ON THE IMAGE QUALITIES OF MAGNETIC PARTICLE IMAGING(Nazarbayev University School of Engineering and Digital Sciences, 2024-05-17) Mukhatov, AzamatToday, scanning methods are getting more popular and becoming an important part of many devices like microelectromechanical systems (MEMS), light detection and ranging (LiDAR) [1], atomic force microscopy (AFM) [2], medical imaging techniques (MRI [3]–[6] and MPI [7]–[11]), and mapping and surveying mechanisms [12], frequency modulated gyroscopes [13]. However, even though scanning techniques have many uses, one of the most important is in medical imaging. These pictures are important because they can be used to see inside the body without needing surgery. They help doctors diagnose, keep track of, stop, and treat many different illnesses [14], [15]. These techniques are used to look at the patient's field of vision and take a picture to study later to understand how the patient is doing. Choosing the right scanning path is very important to get the correct results. By picking the best path, we can scan faster and make the pictures clearer to help diagnose better. This means that the way a scan is done is very important for helping patients [16]. It's important to note that all the mentioned methods are still being worked on by researchers to make them better. Even though the field is getting bigger, the main issue with current scanning methods is that it's hard to accurately estimate the size of the pixels for different scanning settings. For instance, we don't know how big each pixel will be in the scan, with a particular way of scanning a certain area and set of scanning settings. Remember that the size of the pixels you choose will affect how good the image looks after it's scanned. So, it's really important to understand how the scanner moves and works in the area it's focused on, including how dense the scanning is, how much time is spent scanning, the quality of the signal compared to the background noise, and any mistakes in each small area. It is important to think about the right size of the pixels and the space between the pattern and how it is spread out in the FOV. It is important to tell apart the ideas of image resolution and spatial resoution. The sharpness of an image depends on how many tiny dots are in the picture, and how big each dot is. For example, an image with lots of small dots instead of a few big ones will have a clear picture. So, the quality of the image is affected by the size of the pixels. The image resolution decides how much detail and sharpness you can see in the picture. On the other hand, spatial resolution means the smallest detail you can see in a picture, which determines how much detail a camera or sensor can show. Spatial resolution is how small of a thing you can see. It can be measured in millimeters, micrometers, or even nanometers. Usually, to see small details in a picture, the picture needs to have a higher resolution than the spatial resolution [17]. This means the pixels should be much smaller than the spatial resolution. This paper focuses on how the quality of images is affected by the way they are scanned, and the scanning settings used.Any system that scans a particular area has a scanning point that moves in a specific pattern [6], [18]. The quality of the scanned images can change depending on the path chosen, which can also affect how long it takes to scan them and how clear they are [8]. Therefore, it's important for system operators to be able to measure image resolution using pixel size and understand how it's related to scanning parameters [19]. This work aims to create a theory that can figure out the smallest image resolution or biggest pixel size by using all the points where paths cross in the entire view. For a range of paths that may be used in biomedical imaging, the image resolution and its effect on the quality of the reconstructed image are also assessed. These days, a variety of scanning trajectories are accessible, such as spiral, radial, unidirectional, bidirectional Cartesian (BC), triangular Lissajous (TL), sinusoidal Lissajous (SL), and radial Lissajous (RL), as well as different enhanced and unidirectional trajectories. BC, TL, SL, and RL—will be the focus of this thesis because of their high scan resolution, reasonably regular pattern generation, and capacity to provide high-quality reconstructed images with isotropic resolution [8], [20]. Also, the influence of scanning repetition on the quality of reconstructed images in Magnetic Particle Imaging (MPI) systems is thoroughly examined in this research. In order to reconstruct images of various phantoms, were investigated using MATLAB simulations. Simulations were methodically carried out with different numbers of repetitions - 1, 2, 4, and 8 - to obtain a more detailed understanding. The trade-offs between trajectory accuracy, precision, and uniformness were well-explained by this investigation. Using performance indicators such as Normalized Root Mean Square Error (NRMSE), Normalized Total Square Error (NTSE), Peak Signal to Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM), a thorough analysis was conducted in the post-reconstruction phase to compare scanning trajectories. Finding the trajectory that provided the most exact and accurate image reconstruction was the goal.Item Embargo AIRBORNE PARTICULATE MATTER IN ASTANA, KAZAKHSTAN: POTENTIALLY TOXIC ELEMENTS, LUNG BIOACCESSIBILITY, AND RISK ASSESSMENT(Nazarbayev University, School of Engineering and Digital Sciences, 2024-04-26) Agibayeva, AkmaralThe degradation of air quality remains one of the most critical environmental concerns. Exposure to airborne pollutants is extensively associated with various health conditions, including respiratory and cardiovascular diseases, and premature death. The health risks of air pollution have been linked to particulate matter (PM) and its constituents. Potentially Toxic Elements (PTEs) in atmospheric PM are a critical factor contributing to its toxicity. This doctoral thesis addresses multiple aspects of air quality in Astana, Kazakhstan, offering a holistic understanding of the local air pollution situation through (1) analysis of PM and gaseous pollutant concentration; (2) proposing a modification to the toxicity assessment of PM-bound PTEs via in vitro lung bioaccessibility; (3) the assessment of health risk due to inhalation exposure to PM using bioaccessible concentration of PTEs; (4) morphological characterization of PM; (5) source identification; (6) studying precipitation chemistry and its role in air pollution; and (7) assessment of the public knowledge, perception and attitude towards local air quality in Astana. The methodological framework involved primary data analysis (342 PM samples collected in Astana, Kazakhstan from 2021 to 2023) and air pollution data obtained from monitoring stations located in the city (S1-S6) in 2018-2020. Annual and 24-hour mean concentrations of PM2.5, PM2.5-10, and gaseous pollutants (SO2, CO, NO2, NO, and HF) were, in general, higher than established national and international (World Health Organization (WHO)) maximum permissible levels (e.g., for PM2.5 annual mean of 29.7 μg/m3 in 2018-2019; and 24-hour mean of 28.7 μg/m3 (maximum: 534 μg/m3) for PM2.5 and 226 μg/m3 (maximum: 1,564 μg/m3) for PM2.5-10, respectively, in 2021-2023). To simulate real-life inhalation exposure to PM-bound PTEs, the assessment was conducted through optimization of in vitro lung bioaccessibility testing in simulated lung fluids (SLF) (i.e., modified Gamble’s solution (GS) and Artificial Lysosomal Fluid (ALF)). For a modification of commonly established methodology, a large set of PTEs (Cd, Co, Cr, Cu, Mn, Ni, Pb, Sb, V, and Zn) has been investigated using seven distinct formulations of GS, one ALF on two reference materials (SRM 2691 and BGS 102). The bioaccessibility of the selected PTEs generally increased in modified GS with the incorporation of 5% DPPC (phospholipid) (e.g., from 2.87% to 8.35% for V in BGS 102), 0.25% cholesterol (e.g., from 27.3% to 31.5% for Cr in SRM 2691), and 5% DPPC + 0.5% cholesterol (e.g., from 43.5% to 51.5% for Cu in BGS 102). Therefore, using DPPC + cholesterol may be recommended for routine bioaccessibility testing. The effect of the tested solid-to-liquid ratio (S/L) was sample and element-specific. Overall, a lower S/L led to a higher bioaccessibility % in ALF. For all PTEs, the peak bioaccessibility was reached at a 4-week extraction, suggesting a longer testing duration when feasible. The optimized parameters for in vitro bioaccessibility were later applied for inhalation bioaccessibility of selected PTEs (i.e., Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, V, and Zn) in PM2.5 collected in Astana, Kazakhstan. The highest bioaccessible concentration was observed for Fe (mean: 16,229 mg/kg, range: (906-30,419 mg/kg) and V (mean: 10,725 mg/kg, range: (687-27,092 mg/kg). The inhalation Health Risk Assessment (HRA) using a bioaccessible concentration of PTEs in PM2.5 revealed acceptable carcinogenic and non-carcinogenic risks for adult and children exposure, although the maximum Cancer Rate (CR) for adults was slightly higher (1.01 × 10-6) than the established United States Environmental Protection Agency (U.S. EPA) threshold (HIc > 1 × 10-6). Scanning Electron Microscopy (SEM) analysis determined several major PM particle groups, including bioaerosols, coal fly ash (CFA), dust (natural or construction), and soot particles. Irregularly shaped, small-sized particles of CFA are associated with respiratory conditions and neurodevelopmental disorders, while soot particles of complex shapes can penetrate deeply into the respiratory system. In precipitation analysis, the mean concentration of major ions (i.e., F-, Cl-, NO2-, NO3-, SO42-, PO43-, K+, Na+, NH4+, Ca2+, Mg2+) remained within permissible levels for groundwater, drinking, and surface water. However, in April, the highest F- concentration (1.82 mg/L) exceeded the WHO limit for drinking water (1.5 mg/L). The concentration of most heavy metals (i.e., Cd, Co, Cr, Cu, Mn, Pb) was below WHO's maximum permissible levels, except for V, which exhibited the highest average concentration of 108 µg/L in precipitation samples across four seasons. The chemical analysis of PM and precipitation revealed common sources, including coal/liquid fuel combustion and vehicular exhaust. PM2.5 concentration modeling via Multiple Linear Regression (MLR) and Machine Learning (ML) Random Forest (RF) algorithms revealed PM10 and CO as major predictors of PM2.5 concentration. A real-life pollution scenario using Conditional Bivariate Probability Function (CBPF) analysis also suggested a substantial contribution of coal-heated power plant activity (CHPPs) and coal combustion from residential heating, coupled with emissions from internal combustion engine vehicles. Structural equation modeling (SEqM) was employed to investigate the causal relationship between perceived air quality, environmental literacy, and willingness to pay (WTP) for environmental protection. The age, education, and health status of the participants significantly affected (p < 0.001) their level of environmental knowledge and awareness. The SEqM analysis indicates that knowledge is the major determinant in improving public awareness and perception of local air pollution (path value = 0.626). The findings of the current research work can assist healthcare professionals and environmental researchers in public health-related decision-making and establishing feasible air quality guidelines.Item Restricted NANOSTRUCTURED MATERIALS FOR NEXT GENERATION LITHIUM-SULFUR RECHARGEABLE BATTERIES(Nazarbayev University School of Engineering and Digital Sciences, 2023-04-23) Baikalov, NurzhanLithium-sulfur (Li-S) batteries have become increasingly popular as a viable energy storage solution in recent years, providing an affordable replacement to lithium-ion batteries. Li-S batteries have several inherent benefits over LIBs, including a high specific capacity due to elemental sulfur (S8), that when combined with a lithium anode theoretically yields high gravimetric energy densities nearly ten times greater than lithium-ion batteries. However, some issues must be addressed before Li-S batteries can be commercialized, such as low sulfur loading, limited sulfur utilization, and the shuttle effect, all of which contribute to poor cycling stability, capacity loss, loss of active material, lithium anode degradation, and severe self-discharge. The research presents innovative advancements in modifying separators and cathode materials for Li-S batteries. Comparative analysis among nickel, cobalt, and iron revealed that Ni@NGC modification enhances cycle performance and reaction kinetics during lithium polysulfide conversion. Among different modifications, Ni@NGC stands out in various electrochemical aspects, demonstrating stable cyclability and low polarization. Post-cycling morphology analysis reveals a homogenous coating of sulfur compounds on the outer layer of current collectors and separators with Ni@NGC, explaining positive results electrochemical behavior of Li-S battery. Studies highlighted the substantial influence of metal weight percentage in Ni@NGC separator modifications on reaction kinetics and electrochemical characteristics, with 9% exhibiting the optimal ratio for nickel loading to material surface area. Electrochemical performance of separators with 9 wt% Ni loading is improved by capacity retention and reduced polarization. Amongst the separator modifications, the Ni@NGC_9 composite ideally balances the ratio of surface area to Ni content, improving Li-S cell reaction kinetics and cycle performance. After 200 cycles, even with 4.0 mg cm–2 sulfur loading, Ni@NGC_9-modified separator batteries maintain 77% capacity at 0.5 C. A novel cathode current collector, Ni/NiO@CNF, was proposed to expedite sulfur redox kinetics and mitigate the shuttle effect by leveraging the catalytic properties of Ni nanoparticles and the immobilization effect of NiO nanoparticles. Despite the generation of Ni/NiO NPs, carbon nanofibers structure remains mostly unchanged, with improved flexibility. The inclusion of Ni/NiO NPs enhances electron pathways, resulting in noteworthy initial discharge capacities and total coulombic efficiencies at different rates. Polar and catalytic properties of Ni/NiO NPs play a significant role concerning immobilization, facilitating higher kinetics of lithium polysulfides (LiPS) transformation during redox reactions and contributing to an overall enhanced electrochemical performance.Item Open Access ROBUST DATA-DRIVEN PREDICTIVE MODEL FOR BRAIN-COMPUTER INTERFACE(Nazarbayev University, School of Engineering and Digital Sciences, 2024-04-10) Dolzhikova, IrinaA Brain-Computer Interface (BCI) system enables communication and control between a user and an external device without relying on peripheral and muscular activity. Effective control of such a device hinges on accurately recognizing and decoding intricate brain activity patterns generated by the user. The goal of this PhD project is to develop a robust model for predicting human mental intentions using electroencephalography (EEG) signals. EEG, a widely used non-invasive method for monitoring brain activity, is considered due to its ethical considerations, relatively low cost, and its ability to provide a high temporal resolution of received signals. The robustness of the system is verified based on the classification accuracy with respect to the previously unknown subjects such that the performance of subject-independent (SI) BCI system could be evaluated. An essential challenge in BCI research is developing a classifier capable of interpreting users' mental states from EEG data collected from independent subjects. The focus on SI classification is justified because it can lead to BCIs that eliminate the need for individual calibration processes. Over the past few years, deep neural networks (DNNs) in general, and in particular Convolutional Neural Networks (CNNs), have shown impressive training efficiency and performance, leading to the development of state-of-the-art architectures for accurate EEG classification. In this Thesis, to further enhance the performance of the CNN in SI classification, multi-subject ensemble CNN (MS-En-CNN) models are designed. These are the ensembles of CNN classifiers where each base classifier is built using data aggregated from multiple subjects. Based on the distribution of subject-specific data for training and tuning the base learners of the ensemble, three design strategies for MS-En-CNN are introduced: Subject-Specific Training and Model Selection (SS-TM), Subject Pairs Training and Model Selection (SP-TM), and Delete-a-Subject-Jackknife (DASJ) approach. The predictive performance of the proposed techniques is evaluated across two BCI paradigms, namely motor imagery (MI) and P300, using various publicly available datasets. Empirical results show that with any of the presented strategies constructing MS-En-CNN leads to a significantly better SI classification performance with respect to the average performance of the base CNN classifiers. Moreover, MS-En-CNN notably enhances average classification accuracy compared to a single CNN trained on pooled data from training subjects. Among the three strategies, the latter approach, a jackknife-inspired deep learning technique, emerges as the most promising one. It is then benchmarked against state-of-the-art methods, highlighting its superior performance in single-trial SI classification. While these results show potential for datasets with a small number of subjects, addressing computational requirements for large-scale datasets involves extending this approach through the consideration of K-fold cross-validation (CV). In this extended approach, instead of deleting a single subject to form a jackknife sample, a group of K subjects is set aside. On one of the largest MI datasets a K-fold CV-based MS-En-CNN demonstrated a statistically significant improvement (p < 0.001) over the best previously reported results. In addition to MS-En-CNN, proven as a simple yet effective method to enhance the performance of existing CNN models, a new adaptive boosting strategy on the basis of CNN base classifiers (AdaBoost-CNN) with iterative oversampling is proposed. This innovative approach is contrasted with the conventional sample reweighting method, showcasing its potential. Encouraged by promising results, the AdaBoost-CNN warrants further investigation. Overall, this study highlights the effectiveness of MS-En-CNN and AdaBoost-CNN and offers valuable insights that pave the way for further advancements in SI classification within BCI applications.Item Restricted DEVELOPMENT AND OPTIMIZATION OF ML BASED COMPREHENSIVE MODELLING FRAMEWORK FOR GAN HEMTS(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-24) Saddam HusainRadio Frequency (RF) Power Amplifier (PA) is one of the most pivotal constituents of any wireless transceivers. However, continual advancements and ever-increasing complexity in the wireless communication technologies demand frequent innovations in the design of RFPAs. The quality of the designed RFPAs are generally evaluated based around two basic figures of merits namely efficiency and linearity. Thus, the RFPAs should provide maximum power and efficiency while maintaining highly linear operation. In literature, two primary PA design mechanisms, namely measurement- and modeling-based techniques have been extensively utilized. Each class of technique has pronounced merits, limitations and applications. However, owing to the seamless integration ability of the modeling-based techniques with Computer-Aided Design (CAD) tools, they are increasingly becoming more popular. The design and innovation in RFPAs are excessively contingent on the measurement facilities and the Large Signal Models (LSMs) of transistor devices. At present, Gallium Nitride (GaN) High Electron Mobility Transistor (HEMT) technology is regarded as an optimal microwave transistor technology for the design of RFPAs in advanced RF/microwave and high power switching applications. This is due to their attributes namely high energy bandgap, high saturation velocity, high electron mobility, exceptional thermal behavior and high breakdown field. Furthermore, GaN HEMTs manifest high power density, thus a smaller size device can be used to sustain a high power demand. It also implies reduced lower capacitances and lower combining losses in the design of RFPAs and Low-Noise Amplifiers (LNAs). At this point, it is essential to mention that, in general, the available LSMs of GaN HEMTs are very specific and therefore not readily useful for broad range of PA designs. Therefore, there is a pressing requirement to develop accurate, reliable, efficient and robust LSMs of GaN HEMTs which can be readily incorporated in CAD tools. Nevertheless, Small-Signal Model (SSM) development is the first step in pursuit of developing accurate and efficient LSMs. But, both SSMs and LSMs of GaN HEMTs are essential for the design of accurate, efficient and reliable GaN HEMT based RFPAs. Apparently, various modeling schemes have been exploited to develop SSMs and LSMs for GaN HEMTs, however, usually, they are classified into three main groups, which are physics-based, Equivalent Circuit (EC) and Behavioral Modeling (BM) frameworks. This thesis is originated in response to the scientific and technical challenges in EC and BM frameworks for GaN HEMTs at high frequency applications. Among these challenges, the major focuses are on the development of SSMs for GaN HEMTs, which are simple, accurate, computational and time efficient, reliable, scalable, and CAD adaptable. Furthermore, special attention is given to develop SSMs, which manifest strong interpolation and extrapolation abilities. The developed SSMs are then utilized to realize the eventual LSMs for GaN HEMTs. In order to develop SSMs and LSMs for GaN HEMTs, which possess the above-mentioned characteristics, in this thesis, Machine Learning (ML) based approaches have been explored and utilized because of their superior learning, prediction, and extrapolation abilities. However, it is pertinent to state that the ML based modelling of GaN HEMTs is still in its early exploration phase, and various issues related to this type of modelling are unexplored and not thoroughly discussed in literature. It is therefore, in this thesis, an extensive appraisal and analysis of ML and optimization based small-signal and large-signal modelling for GaN HEMTs have been presented. In the first part of this thesis, a detailed comparative analysis of EC based accurate, robust and efficient SSM parameter extraction methodologies for GaN-on-Diamond HEMTs has been demonstrated. For this, initially, a Scanning- Based Systematic (SBS) model parameter extraction approach is developed and applied on GaN-on-Diamond HEMTs. Thereafter, marine predators algorithm, pelican optimization algorithm and tunicate swarm algorithm, the recently developed Optimization Algorithms (OAs), based hybrid extraction methodologies have been developed and applied on the same GaN HEMTs. Finally, a detailed comparison of OAs and SBS modelling schemes by using SBS extraction approach as a benchmark in terms of reliability, accuracy, convergence behavior, complexity, execution time, and scalability is provided and thoroughly discussed. Accurate, efficient and CAD compatible small-signal behavioral models for GaN HEMTs using Artificial Neural Network (ANN), Support Vector Regression (SVR) and Gaussian Process Regression (GPR) based ML techniques have been developed, validated and discussed in the subsequent part of this thesis. These ML based approaches have been applied on many GaN HEMTs devices grown on Silicon (Si), Silicon Carbide (SiC) and Diamond substrates. Furthermore, a meticulous evaluation of ANN algorithms implemented in MATLAB, Python (using Keras, PyTorch and Scikit-learn) and R (using H2O) for small-signal behavioral modelling of GaN HEMTs has been presented. To establish the appropriateness of software environments in distinct application settings, the developed models are examined on a range of metrics namely behavior on the unseen data, training and prediction speed and ADS adaptability, and software environments are surveyed for support and documentation, user-friendly interface, simplicity in the model development procedure, open-access and cost. Optimization of the hyperparameters of ML algorithms is vital to realize the best possible models. In this context, hybrid optimized ML algorithms namely Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) assisted ANN, PSO and RSO assisted SVR, RSO assisted GPR, and RSO assisted various tree-based models are explored and developed. Finally, the developed models are evaluated on many regression tests to identify the most fitting ML algorithms for particular applications. Finally, the last part of the thesis presents ML based CAD adaptable advanced models and applications. Initially, GPR, GA-ANN and RSO-Decision trees based SSMs for GaN HEMTs are developed. Then, the integration of these developed models with ADS are presented by inserting the developed ML based models into a design of class-F PA. Subsequently, to examine the accuracy of the models, stability and gain tests of the GaN HEMT based amplifier configuration are performed. Thereafter, using the developed SSMs, a joint EC-behavioral LSM for a GaN HEMT is developed and presented. The intrinsic drain and gate currents are modelled using GA-ANN, PSOSVR and GPR based approaches. These current modelling approaches are compared in terms of simplicity in the model development stage, computational efficiency, accuracy and required time to simulate the currents. At last, LSM validation and realization using GA-ANN based approach are demonstrated on a design of an inverse class-F PA.Item Restricted MULTI-DISCIPLINARY DESIGN ANALYSIS AND OPTIMIZATION (MDAO) OF WIND TURBINES(Nazarbayev University School of Engineering and Digital Sciences, 2024-01-31) Batay, SagidollaThe purpose of this thesis is to enhance the progress of an open-source platform called DAFoam, focused on Multidisciplinary Design Optimization (MDO). More precisely, the aim is to tailor DAFoam for MDO in wind turbines, serving the wind energy community. Wind energy is becoming increasingly important as a renewable energy source due to its environmental and economic benefits. The wind turbine blades play a crucial role as they directly engage with the wind, exerting a substantial influence on the overall performance of the system. Consequently, optimizing the design of these blades is vital to improve efficiency and minimize costs in wind turbine operations. Low-fidelity simulation and optimization, as well as high-fidelity optimization, are concepts often used in engineering and scientific research, particularly in the field of wind turbine design and analysis. They are interconnected approaches that help engineers and researchers refine and enhance the performance of wind turbines while managing computational complexity and resource requirements. Low-fidelity simulations involve using simplified or coarse models to represent the behavior of a system. In the context of wind turbine design, low-fidelity simulations might use simplified fluid dynamics models or simplified structural models to predict the performance of the turbine. These simulations are quicker to run and require fewer computational resources compared to high-fidelity simulations. However, they sacrifice accuracy and detail for speed. Low-fidelity optimization refers to the process of tuning or improving the design of a wind turbine using the results obtained from low-fidelity simulations. Engineers use various optimization techniques to iteratively adjust design parameters, such as blade shape, tower height, or generator size, with the goal of improving the overall performance of the turbine. Since low-fidelity simulations are computationally cheaper, they allow for a larger number of design iterations to be explored within a given time frame. Conversely, high-fidelity simulations entail utilizing intricate, precise models that accurately depict the behavior of the wind turbine, encompassing intricate details and complexities. These simulations capture more intricate physical phenomena and provide more accurate predictions of the turbine's performance. However, high-fidelity simulations are computationally intensive and may require significant computational resources and time to run. Between low-fidelity and high-fidelity approaches in wind turbine optimization, there is a trade-off between accuracy and computational cost. Low-fidelity simulations are often used as initial screening tools to quickly explore a wide range of design possibilities and identify promising configurations. Once a set of potential designs is identified, high-fidelity simulations are employed to validate and refine the design further. High-fidelity simulations provide more accurate insights into the complex flow patterns, structural dynamics, and other critical factors affecting turbine performance. However, due to their computational intensity, they are usually limited in the number of design iterations that can be explored within a reasonable time frame. Low-fidelity simulations and optimization serve as a way to guide the design process and narrow down the search space before committing significant computational resources to high-fidelity simulations. This two-tiered approach allows engineers to strike a balance between accuracy and efficiency, ultimately leading to the development of better-performing wind turbine designs. This thesis explores the application of low-fidelity optimization using QBlade as a preliminary step toward achieving high-fidelity wind turbine optimization. The initial part of the thesis focuses on the implementation of low-fidelity optimization techniques with QBlade, an open-source software widely used for aerodynamic simulations of horizontal-axis wind turbines. The low-fidelity approach serves as a cost-effective and rapid exploration of the design space, enabling the identification of promising design configurations. After establishing the groundwork through low-fidelity optimization, the primary objective of this thesis is to delve into high-fidelity wind turbine optimization. High-fidelity optimization aims to achieve a more accurate representation of the wind turbine's performance characteristics by considering additional complexities and factors such as structural integrity, aeroelasticity, and dynamic behavior. Unlike low-fidelity optimization, which often relies on simplified models and approximations, high-fidelity optimization takes into account finer details and complexities of the wind turbine design. To attain the high-fidelity optimization, concurrent aero-structural multidisciplinary design optimization (MDO) approach for wind turbine blades is implemented, which considers the interaction between the aerodynamic and structural aspects of the blade and optimizes them simultaneously. The optimization aims to maximize the torque generated by the blade while minimizing its mass. The proposed approach uses DAFoam software for CFD simulation, TACS for FEM simulation, and Mphys under the OpenMDAO framework for fluid-structure interaction between the CFD and FEM. The optimization of wind turbine blade design is undertaken using high-fidelity concurrent multi-disciplinary aerodynamic design optimization, employing gradient-based adjoint solvers. This approach is applied through five distinct schemes utilizing DAFoam.