6746 research outputs found
Sort by
Deep Learning Enhanced Energy Market Prediction: A Robust ARIMAX–LSTM Fusion for Crude Oil Pricing
No full textCrude oil is a highly strategic global resource, and price fluctuations significantly impact nearly all economic sectors. Therefore, accurate forecasting of its prices is essential for better financial stability and decision-making. This study aims to develop a robust model using monthly data from April 2004 to January 2024 to predict the price of crude oil. We propose a novel approach that blends ARIMAX and LSTM models using a weighted combination to leverage the strengths of econometric and machine learning methods. Unlike hybrid models, which are solely designed based on a decomposition-optimization structure, in our model, an explicit ensemble with weights via grid searching is used to enhance the model's flexibility and performance. As ARIMAX is more efficient in dealing with linear relationships and exogenous variables, LSTM performs much better and effectively captures nonlinear patterns and long-range dependence. Weight hyperparameter tuning and cross-validation help reduce the risk of overfitting or underfitting in the model. Our empirical results indicate that the LSTM model provides a powerful forecasting baseline. The weighted ensemble model offers a marginal improvement on the chronological test set, and the Diebold-Mariano test confirms this advantage is statistically significant. Cross-validation reveals the standalone LSTM to be highly robust, highlighting the importance of component model selection. This study contributes to a more sophisticated framework for risk assessment in energy policy by revealing the crucial trade-off between a model's period-specific accuracy and its general robustness. © 2025 Elsevier B.V., All rights reserved
Erosive Wear Behavior of FRP Composite Pipes Under Varying Impingement Angles, Impact Velocities and Flow Directions
No full textFiber-reinforced polymer (FRP) composite pipes are emerging as superior alternatives in sectors such as oil and gas, chemical processing, and aerospace, owing to their high strength-to-weight ratio, corrosion resistance, and design flexibility; however, their long-term durability is susceptible to erosion wear when exposed to abrasive particles. This study experimentally investigates the solid particle erosion (SPE) behavior of filament-wound carbon (CFR/EP), glass (GFR/EP), and basalt (BFR/EP) fiber-reinforced epoxy pipes by ASTM G76-18. Tests were conducted under varied impingement angles (30 degrees, 45 degrees, 60 degrees, 90 degrees), flow directions (axial and radial), and particle velocities (28 and 34 m/s), using both erosion rate (ER) and volumetric material loss to assess performance. All composites demonstrated a semi-ductile erosion response, with degradation consistently peaking at a 45 degrees impingement angle across all test conditions. An increase in particle velocity from 28 to 34 m/s induced a near two-fold escalation in ER. Among the materials, BFR/EP exhibited the highest erosion rates, whereas CFR/ EP was the most resistant. Notably, ER values were consistently higher in the axial flow direction, exceeding radial values by 20-40 % under the most severe condition (45 degrees at 34 m/s). Paradoxically, despite its lower ER, CFR/EP suffered greater volumetric material loss than GFR/EP, a discrepancy attributed to its significantly lower fiber volume fraction (42.4 %) compared to GFR/EP (68.9 %) and BFR/EP (59.7 %). These findings emphasize that both ER and volumetric loss are critical metrics for designing thin-walled pipes, thereby providing a crucial scientific basis for material selection in environments characterized by erosive, multi-directional flow
Determination of Traffic Impact Level in Urban Cycling
No full textIn recent years, the habit of cycling has been increasing. Particularly when active mobility is gaining prominence, there is a global emphasis on healthy living and natural sustainability. Although the current rate of bicycle use in Turkey is quite low, there is significant potential for cycling in metropolitan areas and districts. The concept of bikted (Traffic Impact Level in Bicycle Usage) has been developed to enhance bicycle usage in Turkey, address infrastructure deficiencies, and ensure more comfortable cycling. This method, which consists of parameters related to traffic infrastructure, environmental factors, and user behavior, was evaluated using eight parameters in corridors and five parameters at intersections. In corridors, assessments were made for separated bicycle paths, bicycle lanes, and roads without infrastructure; at intersections, evaluations were conducted for signalized intersections, modern roundabouts, and intersections with traffic markings. In corridors; slope, noise level, curbside parking, vertical marking, surface vibration amount, main road-side road intersection situations, speed limit and bicycle-vehicle gap distance were examined. In intersections; parking at the intersection, intersection visibility, intersection crossing distance, vertical marking presence and bicycle path presence were examined. Additionally, an experimental e-bicycle was developed to aid data collection for bikted. The scoring system in the model was designed using data obtained from field studies and previous literature. For the first time in a bicycle model study, noise intensity, gap distance measurement, slope and vibration were combined for corridor assessment. Slope accounts for approximately one-third of the scoring in corridor assessments for each infrastructure type, and corridors with high slopes cannot reach the \"comfortable use\" classification. The measurements may not be as reliable at intersections as the numerical data analysis conducted for corridors, but they still provide valuable insights for analyzing intersections. This study aims to contribute to the current state of bicycle corridors in Turkey’s traffic infrastructure and future bicycle infrastructure projects, thereby promoting increased bicycle use. Furthermore, bikted is expected to raise awareness among local governments when planning and implementing bicycle-related projects
Cleaning of Fine-Grained Lignite by Two-Stage Hydrophobic Flocculation Using Different Waste Oils
No full textIn this study, the conditions for obtaining clean coal from fine-grained lignite suspensions by two-stage hydrophobic flocculation using different waste oils were investigated. Waste vegetable oil, waste hydraulic oil and waste engine oil were chosen as bridging liquids for hydrophobic flocculation tests. During the studies, sodium silicate was used as the dispersant. Acetone was used at each stage to clean the floc obtained from the agglomeration process. The ash content (%) and combustible recoveries (CR, %) of the floc obtained at the end of each experiment were determined. In addition, contact angle (θ) and calorific values (kcal/kg) were measured and the results were evaluated in detail. At the end of the cleaning stages, low ash clean coal was obtained with a very high combustible recovery. In addition, it was observed that the calorific values increased considerably from 5128 to 5772,5558 to 6304 and 5447 to 5732 using waste vegetable oil, waste hydraulic oil and waste engine oil, respectively. © 2024 Taylor ; Francis Group, LLC
Classification and Analysis of i>agaricus Bisporus/I> Diseases With Pre-Trained Deep Learning Models
No full textALBAYRAK, UMIT/0000-0002-7942-7191; GOLCUK, ADEM/0000-0002-6734-5906; Tasdemir, Sakir/0000-0002-2433-246X; CORUH, Ugur/0000-0003-4193-8401; AKTAS, SINAN/0000-0003-1657-5901This research evaluates 20 advanced convolutional neural network (CNN) architectures for classifying mushroom diseases in Agaricus bisporus, utilizing a custom dataset of 3195 images (2464 infected and 731 healthy mushrooms) captured under uniform white-light conditions. The consistent illumination in the dataset enhances the robustness and practical usability of the assessed models. Using a weighted scoring system that incorporates precision, recall, F1-score, area under the ROC curve (AUC), and average precision (AP), ResNet-50 achieved the highest overall score of 99.70%, demonstrating outstanding performance across all disease categories. DenseNet-201 and DarkNet-53 followed closely, confirming their reliability in classification tasks with high recall and precision values. Confusion matrices and ROC curves further validated the classification capabilities of the models. These findings underscore the potential of CNN-based approaches for accurate and efficient early detection of mushroom diseases, contributing to more sustainable and data-driven agricultural practices.This study has been supported by the Recep Tayyip Erdo ; gbreve;an University Development Foundation (Grant number: 02024011025164).Recep Tayyip Erdogbreve;an University Development Foundation [02024011025164
Thermo-Responsive Shape Memory Behavior of Epoxy With Different Combinations of Curing Agents and Filler Materials
No full textIn this research, bisphenol F-based epoxy composites were prepared with two different curing agents (methyl-5-norbornene-2,3-dicarboxylic anhydride (MNA) and cycloaliphatic polyamine) and polyethylene glycol (PEG) as a plasticizer. Different amounts of sepiolite and colemanite (1, 3, 5 and 10 wt.%) were used as sustainable fillers. The mechanical and thermal properties, as well as shape memory behavior of the composites were investigated by tensile tests, differential scanning calorimetry (DSC) analysis, and bending tests, respectively. The results showed that the tensile strength of the epoxy increased with increasing filler content. The glass transition temperature (Tg) decreased in the presence of PEG and fillers. The epoxy composites exhibited good shape memory performance. The recovery time of all the composites decreased with increasing temperature. The optimum temperature was found to be at 110°C in the case of the shape memory cycle tests. The recovery values (Rr) of the composites cured with MNA and cycloaliphatic polyamine were 99% and 98%, respectively. Ten consecutive shape memory cycles have been successfully conducted for the epoxy composites including sepiolite. © 2025 Wiley Periodicals LLC.Konya Teknik Üniversitesi, KTÜ
A Novel Core-Shell Fe3O4@SiO2/Co-Cr-B Magnetic Catalyst for Efficient and Reusable Hydrogen Evolution From NaBH4 Hydrolysis
No full textThis study presents a novel core-shell magnetic catalyst, Fe3O4@SiO2/Co-Cr-B, engineered for efficient and reusable hydrogen generation from NaBH4 hydrolysis, offering significant advancement in sustainable hydrogen production technologies. The innovation lies in the synergistic integration of a magnetic Fe3O4@SiO2 core with a bimetallic Co-Cr-B shell, which enhances catalytic activity, structural stability, and facile magnetic recovery. Field emission scanning electron microscopy (FE-SEM) revealed a distinctive grape-like morphology resulting from nanoparticle agglomeration, which increased the surface area and active site accessibility. Transmission electron microscopy (TEM) confirmed a well-defined core-shell architecture with a uniform Co-Cr-B shell thickness of 40-50 nm and a consistent particle distribution. These structural features directly contribute to the catalyst's high hydrogen generation rate of 22.2 L gmetal(-1) min(-1) at 30 degrees C with a turnover frequency (TOF) of 2110.61 mol(H2) molcat(-1) h(-1). The catalyst demonstrated remarkable stability and maintained >90% of its initial activity after six consecutive reusability tests. These findings highlight the potential of this catalyst for large-scale hydrogen production and offer a promising route for industrial applications with improved efficiency and durability.Konya Technical University's Scientific Research Projects (BAP) Coordination UnitKonya Teknik niversitesi [221116033]This study was supported by Konya Technical University's Scientific Research Projects (BAP) Coordination Unit (Project No. 221116033)
The Effects of Balance and Strength on Thermal Heatmap
No full text3rd International Conference on Artificial Intelligence over Infrared Images for Medical Applications, AIIIMA 2024 -- 9 November 2024 through 9 November 2024 -- Virtual, Online -- 322299This study examines the effects of balance and strength on the thermal heat map. Participants' thermal temperature values, isokinetic values and balance values were taken in a specific measurement procedure and the relationship between them was statistically analyzed. All processes were carried out under the supervision of a physiotherapist. According to the statistical results we obtained, a significant relationship was found between isometric strength and thermal temperature values in hamstring group muscles. The relationship between isometric force and thermal temperature map of the hamstring group muscles is important in terms of showing that the hamstring muscle group should be prioritized in the analysis of thermal evaluations of basketball athletes. In addition, a significant relationship was found between balance parameters and temperature values in vastus lateralis and ankle. The thermal finding we observed in the vastus lateralis muscle and ankle region in terms of balance parameters, in addition to the evaluation of ankle injury risks of basketball athletes, has raised suspicion that the frequency of knee injuries is adaptively loaded by the vastus lateralis muscle and especially knee injuries occur as a result of this adaptation. From this point of view, it provides evidence that thermal analysis is a useful tool in the analysis of injury mechanisms and injury risks in understanding the most common ankle and knee injuries in basketball. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025
Optimized Size Sorting of Mxene Particles Via Centrifugal Sedimentation: a Practical Approach Using an Empirical Model and Image Processing Technique
No full textControlling the physical, mechanical, and electrochemical properties of MXene-based materials is crucial for their effectiveness in macroscale applications and is closely tied to the particle size distribution of MXene. This study aimed to accomplish dimensional control and sorting of MXene colloids with different particle sizes using centrifugal sedimentation based on an empirical model. Centrifuge time and rotating speed were identified as key parameters and optimized using a mathematical formula generated from the model, considering particle forces in the solution. A novel image processing technique aimed at ease of use was devised to evaluate the separation process, assuring the audience of its usability. The separation efficiencies were measured individually at rotating speeds ranging from 2900 to 6000 rpm. The optimal experimental settings differed between the supernatant and sediment fractions. The maximum separation efficiency was reached at 86% for the supernatant at 3500 rpm for 49 min and 43% for the sediment at 4200 rpm for 34 min, suggesting that supernatant-based separation is more efficient than sediment-based techniques. This study offers a valuable guideline for separating the sizes of 2D materials. Image processing offers scalable particle size measurement, which improves material property control for a variety of applications.This research is financially supported by the Scientific and Technological Research Council of Turkey (TUBITAK) under TUBITAK 1001-The Scientific and Technological Research Projects Funding Program [Grant no: 221M523] and Canakkale Onsekiz Mart University Scientific Research Projects [Project No: FBA-2024-4644].Scientific and Technological Research Council of Turkey (TUBITAK) under TUBITAK 1001-The Scientific and Technological Research Projects Funding Program [221M523]; Canakkale Onsekiz Mart University Scientific Research Projects [FBA-2024-4644
Increasing the Tactical Effectiveness of Precision-Guided Firearms in Group Usage
No full textInfrared thermal imaging is a technique of visualization that captures the heat energy emitted by objects. Unlike systems dependent on visible light, thermal imaging works well in day and night conditions. This capability provides great advantages, especially under environmental constraints such as fog, rain, or snow. Due to its low susceptibility to environmental factors, thermal imaging systems are a reliable solution for critical applications in military and civilian domains. The ability to detect adversaries in heavy fog or in total darkness during military operations improves the outcome of a mission on the field. Thermographic cameras also find extensive use in civilian security screenings, fire detection, and search-and-rescue missions. Despite the wide potential of thermal imaging, its practical implementation on edge devices is fraught with challenges: there is a strong need to optimize performance and energy efficiency. With limited processing power, low energy capacity, and compact hardware, edge devices require innovative solutions, such as the development of energy-efficient algorithms and lightweight AI models. While deep learning-based object detection models improve the accuracy of thermal imaging considerably, they also involve a high degree of computational requirements, hence a delicate balance of performance and energy consumption arises on edge devices. The main contribution of this thesis is the generic framework for developing energy-efficient and high-performance solutions to object detection and tracking problems in thermal imaging on edge devices. Energy consumption, processing capacity, and object detection performance of various edge device platforms such as the NVIDIA Jetson Nano, Xavier NX, Orin Nano, Orin NX, and Rockchip RK3588 are studied. These devices were utilized for the implementation of thermal image analysis using AI-based object detection models such as YOLOX, YOLOv8, YOLOv9, YOLOv10, and Gold YOLO. The accuracy rate, frame per second, and computational cost of the models have been compared. In this study, the newly introduced RSH and RSHAY in this thesis were applied for finding a suitable AI model that can work best for the edge hardware and for selecting the most efficient system across devices. To strengthen performance in real time, there was the optimization of object detection models, with hardware accelerators utilized actively. This model came out as the best combination of edge hardware and object detection model that was efficient in carrying out thermal image analysis with the YOLOv8 PostDas model on the Jetson Orin Nano edge device at an input resolution of 512x512 pixels. The second crucial contribution provided by this paper is regarding the development of the communicating infrastructure that lets edge coordination happen. At the same time, this layer makes object detection propagate on this server for analysis purposes, wherein the contribution has gone for laying down centralized coordination, altogether new approach and thus contributed toward object tracking with facility brought to coordination amidst several associated devices. Hence, facilitating centralized analysis gave room among various devices' coordination through object tracking between numerous devices being able to avail that resource. The conclusions of this thesis contribute to the improvement of the performance of thermal imaging systems while optimizing energy efficiency. The results of the research will provide scalable solutions applicable in a wide range of systems, from battery-operated portable surveillance devices to unmanned vehicles, contributing both to academic research and industrial applications.Kızılötesi termal görüntüleme, nesnelerin yaydığı ısı enerjisinin algılanmasıyla oluşturulan bir görüntüleme yöntemidir. Görünür ışığa bağımlı olmaksızın hem gece hem de gündüz koşullarında etkili bir şekilde kullanılabilmektedir. Bu özellik, özellikle görüşün kısıtlandığı çevresel koşullarda (sis, yağmur, kar vb.) önemli avantajlar sağlamaktadır. Çevresel faktörlerden etkilenme oranının düşük olması, termal görüntüleme sistemlerini askeri ve sivil alanlarda kritik durumlar için güvenilir bir çözüm haline getirmektedir. Askeri operasyonlarda, düşman unsurlarının yoğun sis, karanlık veya diğer olumsuz hava koşulları altında tespit edilmesi sahadaki operasyonel başarıyı artırmaktadır. Sivil alanlarda ise güvenlik taramaları, yangın tespiti ve arama kurtarma operasyonları gibi çeşitli uygulamalarda kullanılmaktadır. Termal görüntüleme teknolojisinin geniş kullanım potansiyeline rağmen, uç cihazlarda uygulanması sırasında karşılaşılan sınırlamalar, performans ve enerji verimliliğinin optimize edilmesini gerektirmektedir. Uç cihazlar, sınırlı işlem gücü, düşük enerji kapasitesi ve kompakt donanım yapıları nedeniyle, özellikle enerji verimli algoritmaların geliştirilmesi ve hafif yapay zeka modellerinin tasarımı gibi yenilikçi çözümlere ihtiyaç duymaktadır. Derin öğrenme tabanlı nesne tespiti modelleri, termal görüntüleme alanında doğruluk oranlarını artırmakta ancak yüksek hesaplama gereksinimleri nedeniyle uç cihazlarda performans ve enerji tüketimi arasında bir denge kurulmasını zorunlu kılmaktadır. Tez çalışması, termal görüntüleme teknolojisinin uç cihazlarda nesne tespiti ve takibi için enerji verimli ve yüksek performanslı çözümler geliştirilmesine yönelik kapsamlı bir çerçeve sunmaktadır. Çalışmada, NVIDIA Jetson Nano, Xavier NX, Orin Nano, Orin NX ve Rockchip RK3588 gibi farklı uç cihaz platformlarının enerji tüketimi, işlem kapasitesi ve nesne tespiti performansları analiz edilmiştir. YOLOX, YOLOv8, YOLOv9, YOLOv10 ve Gold YOLO gibi yapay zeka tabanlı nesne tespiti modelleri bu uç cihazlarda çalıştırılarak termal görüntüler analiz edilmiştir. Modellerin doğruluk oranları, işlem hızları ve hesaplama maliyetleri karşılaştırılmıştır. Bu değerlendirme için RSH ve RSHAY parametreleri literatürde ilk kez bu tez çalışması ile tanımlanmış ve bu sayede hem uç donanım için en uygun yapay zeka modeli belirlenmiş hem de cihazlar arasında en verimli sistemin seçimi yapılmıştır. Gerçek zamanlı performansı artırmak amacıyla, obje tespit modellerinin yapısı optimize edilerek donanım hızlandırıcılar aktif şekilde kullanılmıştır. Jetson Orin Nano uç cihazında 512x512 piksel giriş çözünürlüğüne sahip YOLOv8 PostDas modelinin, termal görüntü analizinde en verimli uç donanım–obje tespit modeli çifti olduğu tespit edilmiştir. Çalışmanın bir diğer önemli katkısı, uç cihazların birbirleriyle koordinasyonunu sağlayan bir iletişim altyapısının geliştirilmesidir. Bu altyapı, uç cihazlarda tespit edilen nesnelerin ana sunucu bilgisayara aktarılmasını ve merkezi bir analiz süreci gerçekleştirilmesini içermektedir. Merkezi analiz sayesinde cihazlar arası koordinasyon sağlanmış ve nesne takibinde yenilikçi bir yaklaşım ortaya konmuştur. Sonuç olarak, termal görüntüleme sistemlerinin performansını artırırken enerji verimliliğini optimize eden tez çalışması, batarya ile çalışan taşınabilir gözetleme sistemlerinden insansız araçlara kadar geniş bir uygulama yelpazesi için ölçeklenebilir çözümler sunmuştur