Theses and Dissertations
Permanent URI for this collection
Browse
Recent Submissions
Item Embargo DEVELOPMENT OF AN UNPLANNED OVERBREAK INDEX FOR OPEN STOPE MINING: ACCOUNTING FOR DESIGN UNCERTAINTY(Nazarbayev University School of Mining and Geosciences, 2024-04-18) Bolegenov, AdilUnplanned dilution can pose a huge burden on the profitability of operations in mines exploiting open stoping mining methods if these are not adequately designed. The empirical stability graph methods are commonly used for these purposes due to their practicability. Despite their merits, experience shows that these graphical design methods can also lead to excessive unplanned dilution which is not surprising since the stability graph method is an approximate design method by nature. This is due to many reasons, for example the unavoidable uncertainties (both epistemic and aleatory) that come with design parameters. Therefore, there is an increasing need for more accurate design tools in the mining industry. Motivated by these aspects, the purpose of this research is to revisit the conventional stability graphs method for open stope design and to propose alternative design tools that could address some of the limitations of the conventional methods. To this end, firstly, a database of unplanned dilution cases collected from sublevel open stoping operations was compiled and the performance of the conventional stability graph method was examined. Next, based on the compiled data, a new overbreak index (DI) was proposed using the Rock Engineering Systems methodological framework. In addition, the effect of aleatory uncertainties on the stability graph method was explored through a reliability analysis. This was accomplished by using the First Order Reliability Method (FORM) to determine the probability of occurrence of unplanned overbreak. The overall results indicated that: the stability graph method yielded accuracies ranging from 9-78%, depending on the ELOS levels; the proposed Overbreak Index highly correlated with the actual ELOS values. In addition, the reliability analysis revealed that the probability associated with the unplanned ELOS varied between 15-100% depending on the rock domains and the wall types. It was concluded that the Overbreak Index and the reliability analysis results could be considered additional tools useful for a more reliable open stope design if uncertainties associated with input parameters of the design must be considered.Item Open Access MODELING COMPLEX RELATIONSHIPS IN GEOMETALLURGICAL VARIABLES: ENHANCING METHODS WITH ACCEPTANCE-REJECTION AND HIERARCHICAL GAUSSIAN CO-SIMULATION(Nazarbayev University School of Mining and Geosciences, 2024-04-16) Kuanyshev, ShingiskhanResource estimation is the basis of efficient and sustainable mining. Accurate mineral resource estimation is critical to optimizing mine planning, minimizing waste, and ensuring the economic viability of mining operations. Traditionally, resource estimation focused primarily on grade, the concentration of valuable minerals in ore bodies. However, the evolving complexity of modern mining requires a holistic approach that includes not only the grade but also the geometallurgical properties of the ore. Geometallurgical properties, which include attributes such as mineralogical composition, texture, and work index, play a key role in shaping mining operations. Understanding and modeling these properties is essential to unlocking the full potential of mineral deposits. In the context of mining, geological complexity often leads to complex non-linear bivariate relationships between different attributes of ore bodies. These relationships can pose significant challenges for resource modeling, as traditional methods such as Principal Component Analysis (PCA) and Minimum/Maximum Autocorrelation Factor (MAF) are ill-suited to handle such complexities. These methods are inherently linear and may fail to capture the nuanced interactions and dependencies within geologic datasets. This research paper presents a new approach to address the complexity of resource estimation in mining, particularly when dealing with non-linear bivariate relationships between geometallurgical properties. The proposed method combines the accept-reject method with hierarchical sequential Gaussian co-simulation. This approach enables the careful modeling of complex relationships (non-linearity) within geological data, leading to more accurate resource estimates and better-informed mining decisions. A case study is presented in which the acceptance-rejection method with hierarchical sequential Gaussian co-simulation is applied to the modeling of two geometallurgical properties, recovery and chalcopyrite. The study shows how this innovative approach increases resource estimation accuracy by capturing non-linear dependencies and spatial variability between these two variables. Due to the hierarchical nature of geological data, the method adapts to different scales of variability, resulting in more realistic and practical resource models. The findings of this research not only highlight the importance of integrating geometallurgical properties into resource estimation, but also provide a valuable solution for solving complex nonlinear bivariate relationships in geostatistical analysis of other regionalized variables. This approach has the potential to revolutionize resource modeling practices in the mining industry, leading to more sustainable, efficient, and economically viable mining operations. As mining continues to face evolving challenges and requirements, it is essential to incorporate advanced techniques such as the accept-reject method with hierarchical sequential Gaussian co-simulation to exploit the full potential of the Earth's mineral resources.Item Restricted ENHANCING AMBIENT ASSISTED LIVING WITH MULTI-MODAL VISION AND LANGUAGE MODELS: A NOVEL APPROACH FOR REAL-TIME ABNORMAL BEHAVIOR DETECTION AND EMERGENCY RESPONSE(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-28) Zhiyenbayev, AdilThe global demographic forecast predicts a surge to over 1.9 billion individuals by 2050, escalating the demand for efficient healthcare delivery, particularly for the elderly and disabled, who frequently require caregiving due to prevalent mental and physical health issues. This demographic trend underscores the critical need for robust long-term care services and continuous monitoring systems. However, the efficacy of these solutions is often compromised by caregiver overload, financial constraints, and logistical challenges in transportation, necessitating advanced technological interventions. In response, researchers have been refining ambient assisted living (AAL) environments through the integration of human activity recognition (HAR) utilizing advanced machine learning (ML) and deep learning (DL) techniques. These methods aim to reduce emergency incidents and enhance early detection and intervention. Traditional sensor-based HAR systems, despite their utility, suffer from significant limitations, including high data variability, environmental interference, and contextual inadequacies. To address these issues, vision language models (VLMs) enhance detection accuracy by interpreting scene contexts via caption generation, visual question answering (VQA), commonsense reasoning, and action recognition. However, VLMs face challenges in real-time application scenarios due to language ambiguity and occlusions, which can degrade the detection accuracy. Large language models (LLMs) combined with text-to-speech (TTS) and speech-to-text (STT) technologies can facilitate direct communication with the individual and enable real-time interactive assessments of a situation. Integrating real-time conversational capabilities via LLM, TTS, and STT into VLM framework significantly improves the detection of abnormal behavior by leveraging a comprehensive scene understanding and direct patient feedback, thus enhancing the system's reliability. A qualitative evaluation showed high system usability results in a subjective questionnaire during real-time experiments with participants. A quantitative evaluation of the developed system demonstrated high performance, achieving detection accuracy and recall rates of 93.44\% and 95\%, respectively, and a specificity rate of 88.88\% in various emergency scenarios before interaction. After the interaction stage, the performance was boosted to 100\% accuracy due to increased context from user's responses. Furthermore, the system not only effectively identifies emergencies but also provides contextual summaries and actionable recommendations to caregivers and patients. The research introduces a multimodal framework that combines VLMs, LLMs, TTS, and STT for real-time abnormal behavior detection and assistance. This study aims to develop a comprehensive framework that overcomes traditional HAR and AAL limitations by integrating instructions-driven VLM, LLM, human detection, TTS, and STT modules to enhance emergency response efficiency in home environments. This innovative approach promises substantial advancements in the field of AAL by providing timely and context-aware detection and response in emergencies.Item Open Access NEURAL NETWORK BASED FILTER MODELING AND OPTIMIZATION FOR 5G AND BEYOND APPLICATIONS(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-26) Serikbekov, ArkhatDesigning high-performance microwave and millimeter-wave filters is difficult because small changes in geometric dimensions and electrical sizes can significantly affect the filter’s characteristic. Typically, in filter design, the initial values of design variables are optimized to achieve the desired performance. In the field of high-frequency RF device modeling, the use of machine learning (ML) through artificial neural networks (ANN) has gained popularity in recent years. Unlike other RF modeling techniques, ANN-based models require training with sufficient datasets to achieve the desired accuracy level. The input data could be the device’s dimensions, while the output could be the S-parameters. Once trained, the ANN-based model can provide EM-level accuracy and equivalent-circuit-level speed. Additionally, it is highly scalable, allowing for the introduction of more input parameters to make the model more versatile and complex. Therefore, the ANN-based model is an excellent option for high-frequency RF modeling compared to other techniques. The main objective of this research project is to develop an AAN that can be used in design of RF Filters.Item Open Access SIMULATOR FOR TIME-SLOTTED LORA NETWORKS(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-19) Abzalova, Assel; Kairatbek, Abylay; Ten, DenisA new simulator for evaluating Time-Slotted (TS) LoRa networks is presented. Traditional testing limitations, due to node scarcity, are overcome by enabling virtual analysis under diverse conditions. The simulator features optimized code, removes unnecessary phases, and implements a dynamic retransmission policy with Reinforcement Learning for improved performance. This user-configurable tool empowers researchers to explore TS-LoRa behavior and optimize this emerging technology.Item Restricted PREDICTING MICROPROCESSOR POWER CONSUMPTION BASED ON HARDWARE PERFORMANCE COUNTERS(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-22) Popov, AlexandrThis thesis presents a foundation for microprocessor power consumption estimation and prediction framework development for ARM-based devices with limited power resources using hardware performance counters. The study introduces a LSTM RNN model to estimate power consumption based on CPU HPC data without evaluation of other hardware events.. This method has a potential advantage for battery-powered embedded systems, where traditional power measurement tools have small efficiency. The research builds upon previous work in the field, highlighting the importance of energy-efficient designs in the growing IoT market. The proposed framework aims to enhance the battery life of portable devices, by helping developers to optimise the software and enabling devices with real-time power management. The model was trained on the dataset collected in idle, video recording, video streaming and audio recording scenarios and evaluated on RMSE performance. Results of the paper suggest that prediction performance of the RNN LSTM model are lacking, but the use of adaptive algorithms in a regression like evaluation have potential to be effective.Item Restricted DESIGN OF ANGLE-INSENSITIVE TRIPLE-BAND METAMATERIAL ABSORBER FOR ENERGY HARVESTING AT MICROWAVE FREQUENCIES(Nazarbayev University School of Engineering and Digital Sciences, 2024-04-19) Kaidarov, SanzharIn the modern world there is an abundance of unused, free ambient energy, which includes energy that is dispersed from wi-fi and internet usage. Harvesting this energy is of importance, as it can provide charge to self-sustainable, low-power application such as IoT devices. In this paper, angle-incensitive triple-band meta-material absorber was designed for resonant frequencies at 2.45GHz, 5GHz, 6GHz with an average absorption efficiency of 80%, 95% and 99% for the angle of incidence θ from 0° to 60° respectively. The simulation was done in CST studio suite, and the simulated absorber shows good absorption qualities. Implemented design was a 9 by 9 unit cell metamaterial grid de- ployed using RO4350B substrate and it showed some agreement with the simulations. Moreover, the design showed a better absorption qualities, covering a wider frequencies range than expected. Experimental absorption is perfect over the shifted frequency bands.Item Embargo FACTORS INFLUENCING PARENTAL PRIMARY SCHOOL CHOICE IN ASTANA(Nazarbayev University Graduate School of Education, 2024-04-15) Karibayeva, AkmaralThe primary purpose of this study is to explore how parents with different incomes, educational backgrounds, and occupations navigate the process of primary school choice and the factors influencing it in the context of Astana, Kazakhstan. Additionally, this research attempts to identify the opportunities and challenges parents experience when deciding on the primary school selection process. The qualitative research study with semi-structured interviews aims to address the following research question and two sub-questions to achieve the study’s objectives: How do parents in Astana navigate the process of primary school choice? 1) What factors influence parental choice of primary school in Astana? 2) What opportunities and challenges do parents experience when selecting primary schools for their children in Astana? Three parents with contrasting socioeconomic and cultural backgrounds whose children studied at three different types of schools in Astana constituted multiple cases of the study. Within these cases, three main common themes emerged: parental involvement, navigating away from mainstream education, and embracing an international curriculum. This study contributes to social inequality research by examining parents’ cultural backgrounds and the interplay of various types of capital (Bourdieu, 1986) in parental decision-making in school choice. The study manifested that a working-class parent was dependent on what public education offered. Upper-middle-class parents greatly impacted their children’s school choice because of their significant cultural capital, making their navigation in the school choice process more like a strategy than a dependence.Item Open Access A Needs Analysis of the Assessed Writing Genres of a 1st Year Undergraduate Engineering Programme(Nazarbayev University Center for Preparatory Studies, 2015) Ketteringham, MatthewIncreasing numbers of non-native English speakers (NNES) are studying at English-medium universities. This increase of students has stimulated the need for EAP instruction, so students can become competent in the discourse conventions of their chosen academic community. The purpose of this research was to carry out a needs analysis (NA) of a 1st year Engineering programme at an English-medium university. A case study approach was used to gain a deeper understanding of the writing requirements of 1st year Engineers and to influence teaching and learning within the School of Engineering (SOE). The methodology used to carry out the NA included genre analysis of institutional artefacts, and interviews and focus groups with faculty and students.