16168 research outputs found
Sort by
Method for cryptocurrencies transaction analysis.
No full textCryptocurrencies are becoming an important part of the economy. As much as a quarter of Bitcoin wallets are involved in illegal activities, almost every second transfer on the network is made by malicious actors. The market is growing in need of regulation of cryptocurrency cash flows, and investigators lack tools capable of performing a detailed analysis of crypto wallets. This paper investigates the possibilities of machine learning for analyzing bitcoin network transactions. The main objectives of this paper are to conduct a review of relevant literature, prepare a dataset, train models to select the best one, and use it to create a tool capable of detecting illegal bitcoin transactions. The created dataset is suitable for training models that would be able to perform transaction analysis in real time using real, unseen bitcoin network data. Three classification algorithms were used (random forest, ADA Boost, XG Boost). During the classification of real transactions, SHAP analysis of the observation is performed, which can explain which transaction features determined the model's decision. The best results were obtained using the XG Boost algorithm, the most important feature being the transaction fee. Using this model, an intuitive tool with a user interface has been created for investigators of financial crimes on the bitcoin network
Research and development of increased steering angle Macpherson suspension.
No full textIn this project BMW E46 front MacPherson suspension is analyzed and geometry properties defined and compared to modified suspension made for drifting. After assessing the main problems of the currently installed front suspension a new design was proposed with better handling characteristics suitable for professional drifting. To properly evaluate the parameters of such suspension a thorough analysis of literature and research papers was made. Defined the main parameters of a suspension such as toe angle, camber, caster and king pin inclination angles. The characteristics of Ackermann steering were analyzed and established that true Ackermann angle differs according to situation and use of the vehicle and driving style. Usually, when a car is driven in a turn, the wheel which is closer to the center of the turn achieves greater angles compared to the other wheel further from the center. In drifting the same turn is approached with the car sliding sideways ant wheels turned the other way, the same wheel closer to the center is now called the outside wheel, which achieves greater angle compared to the other one, thus, true reverse Ackermann achieved by minimizing scrub from the wheels. Because of this, it is necessary to implement adjustability in the suspension geometry. Investigation was made on the negative side effects of suspension geometry, such as bump steer, which occurs if the lower control arm is not parallel with the tie rod, usually after lowering ride height of a vehicle. Problem solving methods defined, such as steering rack adapters and tie rod spacers. After lowering a vehicle, the angle of lower control arm changes which lowers the roll center and for that a vehicle in a turn experiences greater roll because of the greater distance between the vehicle’s center of gravity and roll center. All these negative effects can be eliminated by designing a new knuckle. By creating a 3D model in CAD, original suspension parameters can be defined. The measurements for suspension elements were made by using a laser measuring device and a 3D scanner. After defining the hard points of the suspension elements, calculations in multibody dynamics software „Race Software“ were made to acquire the information about suspension parameters while the suspension experiences vertical movement and steering rack inputs. These parameters were compared to currently used suspension designed for drifting, differences and issues defined. After evaluating the problems with the current suspension setup, a new professional grade drifting suspension design was proposed, which has the ability to quickly change the castor angle without affecting other parameters, achieves maximum available steering angle without any side effects, which occurs at high steering angles. A mechanism, which allows to change camber angle without affecting the king pin inclination angle was implemented which has not yet been available in the market. The new suspension was installed in the BMW E46 vehicle and suspension parameters measured on the wheel aligner at a professional shop. The parameters were compared with the information acquired by CAD and multibody dynamics simulations
Purškiamuoju būdu padengtas ZnO keturbriaunis chemirezistyvinis jutiklis.
No full textThis project explores the fabrication and characterisation studies of spray-coated zinc oxide (ZnO) nanotetrapods and their composite material (g-C3N4/ZnO-T) based chemiresistive sensor for enhanced UV and gas sensing performance. It deals with the chemiresistive UV and NO2 sensing performance of the spray-coated zinc oxide nanotetrapods (ZnO-T) and their composite with graphitic carbon nitride (g-C3N4) for NO2 gas detection. We utilise three distinct signal transduction patterns during each phase of the project. ZnO nanotetrapods are utilised due to their outstanding sensing features, which include a unique three-dimensional morphology, more active sites for adsorption of oxygen, an efficient electron transport mechanism, and improved sensitivity to hazardous gases. The fabrication of the sensor consists of the spray-coating and has benefits, including being a cost-effective, scalable, and less energy-dissipating technique. This deposition method is carried out onto the surface of electrodes of variable configurations having different interelectrode gaps. This provides the tuning of sensitivity, optimisation, and the understanding of electron transport properties. The sensing performance was investigated by exposing UV radiation (500 μW/cm2) and 10 ppm NO2 gas to observe the variations of the current over time. The performance parameters of the analysis comprise response time, recovery time, repeatability, responsivity, and stability. These parameter values exhibit variable values for distinct interelectrode gaps. These findings provide essential information about the surface resistance of ZnO nanotetrapods and their composite with melamine-derived graphitic carbon nitride (g-C3N4) and electron transport sensing mechanism. This research offers the exceptional benefits of the spray-coated chemiresistive sensor based on ZnO nanotetrapods, and due to the chemically stable, scalable, low-cost, and versatile nature, it opens new avenues for applications in advanced nanomaterial sensing technology, environmental pollution monitoring, industrial safety, and medical diagnostics. g-C3N4 and ZnO working together allow for faster charge transport, reduce recombination and make the signal stronger than the noise during operation of the sensor. Because the ZnO nanotetrapods do not break easily, they are highly stable and suitable for real-world use. By using a range of electrode gap sizes, scientists can learn how the microstructure and field distribution affect the material’s response. Besides, using advanced, cost-effective fabrication techniques like spray-coating allows the platform to become small and easily combined with suitable substrate for portable sensor devices. The results hint at the possibility of detecting other target gases by adjusting the surface of the sensor, strengthening future efforts in building multi-gas detectors and smart sensing networks
An enhanced method for secure data deletion in cloud storage.
No full textThe aim of this work is to improve the architecture, based on the existing method, using physical security modules and the secure environment provided by the confidential computing. This master's thesis examines physical security modules, their operations and purposes, data deletion methods and the operation of confidential computing and the security benefits provided. During the research, a new architecture was designed and implemented using the services provided by the „Microsoft Azure” cloud provider. The proposed improved method uses the secure execution environment provided by the „AMD SEV-SNP” processor functionality. This method ensures secure data deletion from the cloud file storage using a „HSM” module-based key manager. Also, a zeroknowledge proof method is used to verify the secure environment and the key generation actions of the „HSM” module. In the experimentation part, the speed of the method was tested. Also, individual operation comparison was made in various environments in order to observe the performance overhead: • In a secure execution environment • In a simple execution environment The advantages and disadvantages of the architectures of the „FADE" method and the proposed method were also compared. The results of the study showed that the proposed method works faster, and also allows ease of use of existing cloud service integrations. Additionally, it was observed that the secure execution environment has a negative impact on the speed of the proposed method, but it is unnoticable to the user. Although the proposed method requires expensive infrastructure resources (HSM-based key manager), it could be used by large enterprises due to the advantages of the proposed architecture
Automatizuotų transporto priemonių įtaka logistikos operacijoms gamybos įmonėje „X“.
No full textThe research uses a mixed methods approach to study the impact of Automated Guided Vehicles (AGVs) on the logistics operations of a tier-1/2 automotive electronics supplier, namely enterprise “X”. Quantitative and qualitative tools are utilised to explore the impact of the implementation on the performance of plant operations. Logistics processes are mapped to show the scale of AGV implementation, and the preparations made for the robots' systematic integration and future scalability are presented. A maturity model is built based on industry practices discussed in the literature and company expectations for the automated solution. The model assesses implementation across six dimensions: system integration, autonomy, data and KPIs, workforce skills, efficiency, and AI/ML use. Expert professionals from the company are surveyed to diagnose the maturity stage of the AGV implementation. The maturity stage identified was “Standardised”. The strengths of the case company were in defining and tracking Key Performance Indicators (KPIs), while the weaknesses were in efficiency and AI/ML adoption. A quantitative statistical analysis was performed for the following KPIs: quality rate, utilisation rate, effective utilisation rate, performance rate, and total effective equipment performance (TEEP). Control charts that confirm system stability were graphed to identify that the underutilisation of the fleet was the limiting factor rather than technical performance, which was further confirmed by studying hourly average TEEP rates and performing correlation analysis. The economic section calculates the return on investment for the AGV fleet, yielding a positive IRR of 8%. The findings indicate that AGV deployment improves operational efficiency and provides scalable benefits at the operational and economic levels. The project presents a road map for industrial AGV/AMR systems implementation, using applied research methods to extract operational insights and assess maturity levels
Road scene segmentation system.
No full textThe precise detection of objects and the surrounding environment is one of the essential requirements for applying computer vision systems in the automotive, robotic, and aviation industry. The quality of such systems has a strong influence on the actions of autonomous vehicles on the road. To avoid traffic accidents, autonomous vehicles must accurately identify other road users and obstacles. Due to the wide variety of objects and environmental conditions encountered on the road, image segmentation remains one of the most challenging tasks in the field of computer vision. This thesis examines computer vision models for image segmentation, their architectures, and the data sets used, with the aim of creating an efficient system to segment road scene imagery. Based on an analysis of the most effective existing models, an image segmentation algorithm built upon a transformer-based neural network architecture is implemented. In the experimental section, modifications applied to the system are described and compared, and their impact on the accuracy of the image segmentation system is assessed
Strengthening interinstitutional collaboration by consolidating personal healthcare treatment resources at the Marijampolė municipality Health Centre.
No full textThe healthcare system in modern society faces continuously growing challenges – rising patient expectations, limited financial resources, shortages of human resources, and the need to ensure the provision of quality services for all citizens. In this context, strengthening inter-institutional collaboration becomes increasingly relevant, particularly when municipal-level initiatives for consolidating healthcare resources are implemented. Consolidation is a new approach that involves merging or coordinating human, infrastructure, informational, and financial resources from different institutions to provide personal healthcare services more effectively. The relevance of this topic is driven by global trends and the efforts of European Union (EU) member states to improve the quality, accessibility, and efficient use of personal healthcare services (Hearld & Westra, 2022). In Lithuania, since the end of 2023, integrated healthcare systems have been introduced, emphasizing the establishment of Health Centers in municipalities to ensure more efficient resource allocation and better fulfillment of patient needs (Ministry of Health, 2024). However, there is still limited scientific literature addressing the establishment and management of consolidated Health Centers (Keast, Brown, & Mandell, 2007). Therefore, this final project examines the possibilities for strengthening inter-institutional collaboration through the consolidation of personal healthcare treatment resources in the Marijampolė Municipality Health Center
Biological cells recognition and their functional properties evaluation using artificial intelligence methods.
No full textNowadays medicine, biology research has a lot of advanced technologies but one thing that is always slowing down the advancement of them is big counts of data proccessing, simmilarity findings, conclusion and relationship creation. To speed up the reasearch and find new relationships between the data – we can use artificial intelligence methods. The aim of this project – using artificial intelligence automate, speed up biological cells research, with the main goal of cell‘s gap junctions. To detect them, calculate them and evaluate the quantitive parameters. Artificial intelligence methods can detect alive cells but the rapid advancement in the field creates challenges when it is need to cross-compare models, to select most optimal one for the current task at hand. To choose the best model – cross-comparison is done between standard U-NET model vs „Detectron2 RCNN50“ vs YOLOv11. The research results show that „Detectron2 RCNN50“ achieves the highest performance, which are further improved by doing additional post-processing on results. The created cell segmentation and data extraction architecture marks the cells and calculates their counts in pictures. Using segmentation data it is capable of extracting cell quantitative parameters and as such using it is possible to calculate and plot cell‘s hemi-channels permeability scores after transfection, giving the ability to compare transfected cells data with non-transfected cells
Investigation of necessity of cross-border electricity interconnections in Baltic sea region to achieve decarbonisation objectives.
No full textIn order to find out what the development of cross-border interconnections should look like in the light of decarbonisation targets and countries' emission reduction strategies, a study on the development of cross-border interconnections was carried out. This work analysed documents and scientific articles dealing with the transition to 100% RES production and the need for cross-border interconnection development. The paper analyses the energy systems of 9 countries in the Baltic Sea region (Lithuania, Latvia, Estonia, Poland, Germany, Denmark, Sweden, Finland and Norway)
Research of renewable sources impact on transmission network frequency.
No full textThe Master's thesis investigates the influence of renewable energy sources, battery energy storage systems and synchronous compensators on system frequency. This thesis is structured in three parts. The first part presents a literature review which discusses the impact of renewable sources on system frequency. It draws on the literature to identify the main causes of frequency instability and the solutions to these problems using battery energy storage systems. The second part describes the system model under investigation. Scenarios to investigate the system parameters are presented. The third part presents and analyses the results obtained. The results are compared with each other and the impact of RES, BESS and SC on the grid characteristics is analysed. In the final part, conclusions of the study are drawn based on the results and data of the previous parts