http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/171 2026-04-23T13:41:14Z 2026-04-23T13:41:14Z Salmanov, Farid http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1595 2026-03-17T05:43:10Z

Determining the Location of a 220 KV Overhead Line Fault Using the Wave (Traveling Wave) Method Salmanov, Farid This thesis presents a procedure for the accurate computation of 220 kV overhead transmission line fault location using the traveling wave (TW) method of fault location. Compared to traditional impedance-based methods with high sensitivity to fault resistance, load flow, and system parameters, the traveling wave method is highly accurate based on the physical electromagnetic transients propagation time. The traveling wave technique depends on the observation of the high-frequency voltage and current transients created during fault, traveling along the line nearly at light speed. The accurate location of a fault can be computed by measuring the time-of-arrival of the first wavefronts at the line terminals. Essential to such precise high accuracy is sub-microsecond resolution offered by cutting-edge GPS time synchronization and MHz-range high-frequency data sampling that avoid the limitations of earlier TW systems. The research makes use of the two-ended fault location technique, which sets the first wave front arrival times at both local (tL) and remote (tR) line ends. The core of the research is the Simulink model developed in MATLAB to simulate 220 kV transmission line using the distributed parameter model. This model has a specific Traveling Wave Fault Locator (TWFL) subsystem for calculating propagation velocity from line inductance and capacitance and end fault distance. This simulation confirms the high accuracy of the principle, where it is able to identify faults with a precision of several hundred meters and contribute towards quicker restoration of systems and lower outage duration. The work acknowledges significant challenges, including expense and complexity of the high-speed data acquisition equipment required, and possibility of signal noise and wave reflection leading to inaccuracy. Future research directions, like combining machine learning with hybrid TW-impedance schemes, are mentioned as one possible means of further enhancing this significant power system protection technology's reliability and application.

Mammadzada, Raman http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1593 2026-03-16T08:10:14Z

Voltage Regulatiin in Generating Networks Mammadzada, Raman To keep the voltage level acceptable is the primary phenomenon of electrical power system and aiming to keep the voltage around acceptable level different load and generation conditions. Acceptable voltage regulation makes power quality, system reliability, and efficiency better [1]. During the long time, the tap changers have been used to regulate voltage in transformers, and capacitor banks are utilized for regulation of reactive power in electrical power systems. Tap-changing transformers: To adjust voltage dynamically in transformer on-load tap changers regulate transformers turn ratios. Shunt capacitor banks: For supporting local voltage and to make power factors better the capacitor banks are explicitly used. Synchronous Condenser: generate reactive power supply and system consistency, crucial for huge- scale system. These ways, in cutting-edge power system, reliability, not including the speed and flexibility is mandatory.[4] The including of distributed Energy sources and connection of them electrical system can cause some problems related with challenges in voltage regulation. At those times mentioned traditional regulation methods were insufficient for electrical grids and high integrated Distributed energy sources.

Abilov, Shamkhal http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1591 2026-03-16T05:54:09Z

The Diagnosis of Foundations of 220-500 kv Electrical Transmission Line Supports Abilov, Shamkhal This thesis explores a vibration-based diagnostic method for assessing foundation condition in 220-500 kV overhead electric transmission towers. Research evaluates model for the mass-spring-damper system for a transmission tower and its foundation, with realistic values for its mechanical properties obtained from literature studies, coupled with information from the Aghsu 220kV Corridor. Simulation studies are conducted dynamically, with soil properties being varied for different forces, simulating operational forces like wind excitation. Outputs simulated for transient response to these forces include displacement, velocity, and acceleration, processed with filtering algorithms, Fast Fourier Transforms, spectral analysis, and features extracted for diagnostic indicators like frequency deviations, damping ratios, and acceleration magnitude levels, in combination forming a diagnostic foundation to ascertain differences between a healthy foundation from a deteriorated foundation.Simulation results are validated with model predictions based on operational expectations, along with published values from international studies. The simulation results for natural frequency (≈2-12 Hz) and damping coefficients (0.03-0.08) for modelled soil-structure interaction dynamics match published values from observed practices in the energy sector. It has now been validated from simulation analysis that vibration diagnostics have proved to be an efficient, non-destructive technique for condition analysis to monitor Azerbaijan’s 220-500 kV high-voltage transmission network. The thesis explores a vibration diagnostic technique to analyze the condition of foundations for 220-500 kV overhead electrical transmission towers. Conventional visual analysis, adopted in Azerbaijan’s energy power network, explored on-site defects, such as deviations in overhead wires, gaps in support foundations, with no information regarding changes in soil resistance for support foundations. By overcoming these defects, researchers developed simulation-based analysis on mathematical models for dynamic soil-structure interaction principles to explore early indicators for loss, degradation, and damage to electrical support foundations. Starting with creating simplified mass-spring-damper models for the transmission tower, along with its foundation, based on real-world, realistic model parameters extracted from literature, in addition to simulation analogues for the Aghsu 220 kV route, dynamic simulations with varying soil properties for different levels of stiffness, damping, and forces are employed for simulating operational forces such as wind-borne vibrations. Along with simulation results on displacement, velocity, and acceleration, post-processing is performed on these parameters for filtering, spectral analysis based on Fast Fourier Transforms, and indications such as natural frequency deviations, damping ratios, and Root Mean Square acceleration indicators. Outputs from the model are validated by comparison with analytical results and published values in international studies. The match between model results for natural frequency ranging between 2-12 Hz, with damping coefficients between 0.03-0.08, is accurate enough to validate model performance in simulating real-life dynamic performance for transmissions in real foundations. These results validate the utilization of vibration analysis for diagnosing real foundations, particularly for identifying loss in foundation stiffness, which will not be detected during conventional analysis. Concluding, this method is an efficient, non-destructive, and scalable process for testing Azerbaijan's high-voltage power transmissions, with applications for incorporating automated analysis via machine learning algorithms in its evaluation method.

Garashov, Vagif http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1590 2026-03-16T05:42:27Z

Sustainable Solutions in the Transition to Green Energy: Economic and Environmental Perspectives Garashov, Vagif Transitioning to green energy - becomes important every year. The world and humanity are developing fast. More emissions releasing to atmosphere and harming the environment. And conventional energy sources will end one day. That’s why countries aim to integrate into renewable energy sources sooner. This research studies the sustainable steps for integration to clean energy. In addition, regions need strong transmission lines to move clean energy. So, thesis also focuses on HVDC. Azerbaijan’s location between Europe and Asia is very strategic. The new interconnection model was proposed and simulated. The new Central Asia → Azerbaijan → Türkiye → Europe interconnection between regions. The 2030 situation model was simulated for four regions. A 5 GW HVDC line was tested and estimated losses using typical HVDC values were calculated. The results of this model show that Central Asia has extra renewable power. Which can be transferred through South Caucasus by corridor to Europe. Also results show the HVDC losses are very low. And Corridor technically worked at full capacity with benefits. To sum up, the new interconnection model shows that the idea is technically possible. This corridor can support clean energy goals in the region. Additionally, corridor can be expanded in future. And this study is useful base for more detailed studies.

Rzayev, Kamran http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1534 2025-11-10T16:24:45Z

Monaural Multi-Speaker Speech Separation in Azerbaijani Language Rzayev, Kamran The objective of speech separation is to distinguish and separate different speakers’ utterances from each other and often also from background noise. Speech separation is one of the fundamental problems of signal processing domain that has a big diversity of applications, including hearing prosthesis, mobile telecommunication, and robust automatic speech and speaker recognition. The capability of the human hearing to isolate one sound source from a combination of several sounds from multiple sources is exceptional. Humans appear to be capable of following the utterances of one speaker in the presence of other speakers and background noises with extremely low to no amount of efforts even in noisy environments such as parties. The problem of separation of multiple speakers from noise into separate utterances is referred as “cocktail party problem”. This problem has been investigated worldwide but there is no evidence of using Azerbaijani dataset to solve this kind of problem. This paper aims to investigate how different language affects the solution if this problem and proposes several solutions as well. The first approach will use included Support Vector Machine, Multi-Layer perceptron, Decision Tree Classifier, Random Forest Classifier and K-Nearest Neighbors. Additionally, pretrained models will be used to experiment on custom dataset as an alternative to proposed solution. All the models are going to be evaluated on several evaluation metrics such as accuracy, SNR, SDR, precision, recall.

Naghizade, Elshan http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1529 2025-11-05T07:40:27Z

Autoencoder-Based Efficient Feature Extraction Method for Position Invariant Image Naghizade, Elshan This master’s thesis explores an effective method for extracting features from images that do not change position, using a technique based on autoencoders. The focus is on using Convolutional Autoencoders (CAEs) and Variational Autoencoders (VAEs) to process two specific types of data: the MNIST dataset, which includes images of handwritten digits, and the Bristol-Myers Squibb – Molecular Translation dataset, which contains images of hand drawn chemical structures. The approach involves training both CAE and VAE models on these datasets to create latent vectors—compressed representations that summarize key aspects of the images. For the MNIST dataset, these vectors are used to train deep neural networks (DNNs) for classifying the digits. The results show high accuracy, with CAE-based DNNs achieving 96% and VAE based DNNs achieving 91% accuracy on the training set. The performance is further analyzed for each digit, using bar graphs and confusion matrices to demonstrate that CAEs provide more accurate and consistent classifications. For the chemical structure images in the Bristol-Myers Squibb dataset, the thesis tests three different setups for converting images into textual formulas. These setups include a combination of EfficientNet, Vision Transformer (ViT), and a traditional Transformer model, as well as pipelines using features from VAEs and CAEs paired with ViT and Transformer. The effectiveness of these models is measured using the Levenshtein distance, a metric that quantifies the difference between the predicted text and the actual formula. The CAE-based model outperforms the others, showing it can more accurately translate images into text. The study demonstrates the value of using features specific to the data being analyzed. It shows that CAE and VAE models, when trained on specific datasets like the Bristol-Myers Squibb, can capture essential details better than more general models. The thesis wraps up by suggesting directions for future research, such as implementing more advanced data augmentation techniques, applying transfer learning from specialized fields, exploring different neural network structures, testing the models on various datasets, and finding ways to make training and using these models more efficient.

Aliyev, Musadiq http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1528 2025-11-04T11:34:46Z

Optimizing Cash Management in ATM Networks using Time Series Models and Spatial Analysis Aliyev, Musadiq Effective cash management across ATM networks is essential for banks to maintain optimal service levels. However, in ABB, the largest state bank of Azerbaijan, cash management is conducted manually. The main objective of this thesis is to develop an optimal strategy for ABB’s ATM cash demand forecast by integrating time series models and spatial analysis. The significance of this study is that, with the help of a robust strategy, the bank will reduce operational costs, improve service levels, increase customer satisfaction, and enhance profitability. Cash demand forecasting experiments were conducted on ABB’s ATMs to achieve those goals. Various methods were used to build eight different models. Different time series, machine learning models, and Meta’s Prophet model were used. The best-performing model was SARIMAX, with SMAPE score of 5.89% and RMSE score of 83537. Next, spatial analysis was performed by grouping ATMs into five distinct sub-regions based on their distance from the city center. This analysis was done for Baku city to consider the various spatial factors influencing cash demand in different locations. Then, one ATM from each region was selected, and the cash demand for the last nine months was forecasted using SARIMAX. The MAPE score spans between 4.7% and 6.6% which demonstrates that the model is reliable even with different locations, populations, and atm distributions. The forecasted MAPE scores will be utilized as model calibration factors to adjust the fluctuations of the model's predictions in production. The integration of forecast results with spatial analysis to formulate strategies represents the novelty presented in this thesis.

Bashirov, Sokrat http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1525 2025-11-04T06:54:14Z

Azerbaijan Text Clustering using Machine Learning Methods Bashirov, Sokrat In this digital era, the explosion of textual data is causing us to develop sophisticated text mining and clustering methods. Although the state of art has improved for most well-resourced languages, relatively little research had been carried out on a language with smaller resource like Azerbaijani. In this thesis I investigated using clustering algorithms to enhance the information and communication access in Azerbaijani speaking community. 15,500 news articles were used compiled as a part of oxu.az. So, K-means, Fuzzy-Kmeans, Agglomerative Hierarchical Clustering, Spectral Clustering along with Gaussian Mixture Model (GMM) and Latent Dirichlet Allocation were deployed. They were evaluated on the basis of Silhouette Score (SS) and Davies-Bouldin Index. Word2Vec embeddings yield higher ARI than TF-IDF, while Spectral Clustering and LDA report superior scores owing to their capability of mapping complex workout nodes. The future works will improve the Pre-processing, hybrid Clustering and Deep Learning Embeddings. Applications to real-world problems ranging from recommendation systems and content categorization, all of which will build experience with the models.

Murshudlu, Huseyn http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1523 2025-10-30T06:07:06Z

Power Grid Planning and Operation Considering Electric Vehicle Proliferation Murshudlu, Huseyn In integrating electric vehicles into the power grid, a new milestone will be achieved towards the worldwide change in sustainable energy systems. Apart from altering the face of ground transportation, electric vehicles are foreseen to play a crucial role in modern power networks for enhancing grid flexibility, increasing renewable energy use, and thereby reducing green gas emissions. These end-consumers, electric by nature, are potentially units of energy storage that turn out to be the main contributors in the challenge of renewable intermittent and peak demand management. The research proposes a holistic approach to assess the integration of EVs into the power system of Azerbaijan. The study considers multiple electric vehicle penetration scenarios, user charging behaviors, infrastructure constraints, and renewable energy integration. The research work, based on ETAP software program and simulation-based approaches, investigates power flow, demand profiles, and voltage stability for the years 2030, 2040 and 2050. The results indicate remarkable challenges: overloading of the grid and voltage instability after 2040, especially in medium and low-voltage systems, with an estimated increase in peak demand of 523.2 MW by 2040 under high-penetration scenarios. The key results from this analysis include: (a) critical infrastructure constraints will begin to emerge from 2040, and there is a strategic need for investments in transmission and distribution networks; (b) the potential of demand-side management and V2G systems is necessary for maintaining stability on the grid; (c) clean energy sources such as solar and wind can contribute much to sustainable electric vehicle charging; (d) more robust public policies and financial incentives will be required for developing infrastructure readiness; and (e) the integration of variability in user charging behavior and in renewable energy availability needs consideration in long-term planning. In addition, this study is in line with the commitments of Azerbaijan during COP29 through providing actionable strategies toward a robust and sustainable energy system supportive of widespread EV adoption at lower greenhouse gas emissions. The methodology and findings give a foundation to the policymakers and industry leaders in Azerbaijan and other countries facing energy challenges.

Zulfigarov, Asgar http://dspace.ada.edu.az:80/xmlui/handle/20.500.12181/1522 2025-10-30T05:35:53Z

Voltage Regulations in Electrical Systems Zulfigarov, Asgar Voltage regulation challenges are important from both perspectives: the stability of the grid and reliability of power delivery, especially in modern electrical grids that are facing a rapidly increasing share of renewable sources. In this thesis, challenges and possible solutions to voltage regulation issues within the context of the Azerbaijani power grid will be discussed by focusing on advanced technologies of reactive power compensation, hybrid systems, and smart grid automation. Thesis develops and employs a simulation approach using MATLAB to assess voltage-regulating devices and methodologies against peak loading conditions, intense renewable integration, and extreme weather events. The result highlights the importance of sophisticated devices like Static Synchronous Compensators (STATCOM) and Static Var Compensators (SVC) to enhance grid stability. The STATCOM was far more effective at Baku City and other metropolitan areas, reducing voltage recovery times by 40-60%, while SVCs provided economical stabilization for more rural areas like Gobustan. Also, hybrid renewable energy systems that incorporate wind and solar power in synergy with battery storage reduce voltage variations to levels that maintain grid stability even for 70% or higher penetration of renewable energies. The findings point out the absolute need for adaptive approaches to manage such typical fluctuations in renewable energy sources. This underlines modernization research on infrastructure and the use of smart grid technologies in order to tackle the particular challenges Azerbaijan is facing. In fact, long transmission lines connect far-flung renewable energy projects to metropolitan areas and increase voltage instability, creating the need not only for improvement in transmission infrastructure but also for integrating distributed energy generation methodologies. In some cases, smart grid technologies are capable of increasing voltage regulation efficiencies up to 20-30% through real-time monitoring and adaptive control strategies enabled by SCADA systems and machine learning algorithms. These economic analyses confirm, on the whole thesis level, investments in voltage regulation technologies that are feasible both technically and economically. In more detail, all STATCOM and SVC systems have less than one year payback time, while hybrid ones-as Tesla Megapack-could achieve yearly savings up to $25 million annually due to the reduction of renewable energy curtailments. Therefore, automation in smart grids contributes toward better economic and operational efficiencies, which will in turn help Azerbaijan reach its renewable energy targets without conceding on grid stability and reliability. The thesis concludes with actionable recommendations for phase-by-phase deployment of advanced voltage regulation technologies, centered on urban- and rural-specific approaches, hybrid renewable systems, and policy frameworks for leveraged private investment. Further research efforts are thus suggested in order to: continue regional case studies, refine methodologies adopted for the simulations, and review the consistency of national policies with global best practice in terms of promoting a resilient and sustainable energy system. The thesis is supposed to develop a holistic approach to updating Azerbaijan's electric grid, considering the gap between global trends and local needs. The proposed methodology will be oriented in correspondence with the country's priorities in the field of renewable energy sources for the purpose of promoting its goals of energy security, economic growth, and ecological sustainability.