In this paper, an effective method for accurately classifying Electroencephalogram (EEG) data for the early identification of epileptic seizures is presented. The suggested process essentially hybridizes several statistical data, discrete wavelet transformations (DWT), machine learning algorithms, and feature selection techniques independently. The automated multi-resolution signal processing approach decomposes EEG signals into detail and approximation coefficients after splitting them into detailed parts with varying window sizes using DWT. Statistical latent features are extracted from these coefficients that describe the nonlinear and dynamical patterns in the signals. Feature selection techniques were used to reduce the dimension of the feature matrix while highlighting the important elements. Different classifier structures were developed to classify input matrices. For all classifiers, the optimal hyperparameters were found using grid search techniques. Performance metrics for classification were calculated to assess the model's performance. Also, the most important frequency bands were detected to distinguish EEG signals. In the analysis, to compare the proposed procedure with the other approaches in terms of detecting the epileptic behaviors correctly, a benchmark data set from the University of Bonn database was used. The results showed that the proposed approach can estimate more robust models concerning performance metrics and information criteria in classifying EEG signals.
Today, increasing energy costs and environmental concerns direct industrial facilities to sustainable energy sources. In this context, PV Power Plants (PVPP) installed on the roofs of factories offer an environmentally friendly and economical energy production solution. Rooftop solar power plants allow factories to utilize their existing rooftop areas, while meeting a significant portion of their electricity needs and providing significant savings on energy bills. It also contributes to environmental sustainability by reducing carbon emissions. However, there are some difficulties in system installation, such as the suitability of the roof structure, initial investment cost and bureaucratic processes. Considering all these factors, rooftop solar power plant applications have become a strategic investment for factories due to their long-term economic benefits, energy supply security and environmental gains. This study aims to investigate the energy saving performance and environmental impacts of a PVPP installed on the roof of a factory in Düzce, Türkiye. The results show that in the months when solar radiation is intense in Düzce, the electricity used from PVPPs is higher than the electricity used from the grid. However, an average of 46.02% of the energy needed in the factory for 8 months was provided by the PVPP installed on the roof. From an economic perspective, the payback period of the installed solar power plant system is calculated as approximately 3 years. From an environmental perspective, with the installation of the PVPP, CO2 emissions have decreased by 1,100 tons compared to the past. This value means that the factory emits 76% less CO2 compared to the factory before the installation of the PVPP. As a result, it is seen that the PVPP placed on the roof of the factory contributes significantly to economic and environmental aspects.
In classrooms, where the learning process takes place and which cater to many users, indoor visual comfort and energy efficiency are essential for user health and academic success. Ensuring the homogeneous distribution of daylight indoors in classrooms is a significant problem that must be addressed during the design of educational buildings. To address this issue, designers use artificial elements such as shading and louvres in the building envelope. However, producing these elements results in carbon emissions, and they do not decompose naturally once their lifespan is over. Considering the United Nations Sustainable Development Goals (UN SDG) target of achieving net-zero emissions in cities by 2050, it is necessary to minimise emissions from these elements. Within this scope, the study aims to highlight the importance of landscaping in educational buildings to control the amount of natural light entering indoor spaces. Within the scope of the study, two identical classrooms at Balıkesir University Faculty of Architecture were examined. While there is no landscape area in front of the northeast-facing D1 classroom, there is a landscaped area with vegetation in front of the southwest-facing D2 classroom. The study examined the distribution of daylight in these classrooms during the day. It evaluated the effects of the landscape on the daylight performance of the interior spaces by conducting on-site point measurements with a light meter at 9:30, 12:30, and 15:30 on 21 June, the summer solstice. According to the measurement results, the northeast-facing D1 classroom had relatively high illuminance near the windows in the morning. In contrast, the D2 classroom had plants in front of its windows and had a homogeneous daylight distribution throughout the day. The study highlights that optimisation methods can be used in future studies to control the amount of daylight entering the interior of educational buildings through landscaping.
Flow separation and wake formation around circular tubes are among the primary causes of pressure losses and limited heat transfer performance in tube-bank heat exchangers (TBHEs). In this study, the thermo-hydraulic performance of a staggered circular tube-bank heat exchanger enhanced with a novel Inward Curved Ring-Winglet (ICRW) configuration was numerically investigated. Unlike conventional external fins that primarily increase surface area, the proposed intrusion-type design modified the core flow by partially penetrating into the channel region, promoting longitudinal vortex formation while suppressing wake recirculation. A three-dimensional steady-state CFD framework was developed in ANSYS Fluent using the RNG k–ε turbulence model to analyze airflow and heat transfer characteristics. The effects of four geometric parameters, namely winglet length (L), winglet gap (G), inclination angle (θ), and channel height (H), together with the Reynolds number (Re), were systematically examined using Response Surface Methodology (RSM). A Central Composite Design–based RSM framework was employed to construct surrogate models and identify the optimal design by maximizing the thermo-hydraulic performance factor (TPF). The performance evaluation was based on the TPF, which accounts for both Colburn j-factor and friction factor. The investigated parameter ranges were L = 12.5–22.5 mm, G = 0.75–2.25 mm, θ = 3.75°–15°, H = 3.125–12.5 mm, and Re = 1100–11500. The RSM analysis identified an optimal configuration at L = 22.105 mm, G = 2.10 mm, θ = 5.12°, and H = 3.14 mm, for which the maximum TPF of 1.53 was achieved at Re = 11239. Compared to the baseline tube-bank configuration, the optimized ICRW design significantly enhances heat transfer while maintaining acceptable pressure losses. Flow visualization results indicate that the improvement is mainly attributed to intensified longitudinal vortex structures and effective disruption of thermal boundary layers. The results demonstrate that intrusion-type ICRW fins provide a compact and effective passive enhancement strategy for high-performance air-side TBHE applications.
Earthquakes have always been a great danger for humanity. There have been and continue to be important developments throughout human history to minimize the impact of this great danger. Turkey is located in a geologically active region and is home to numerous fault lines, including three major ones. Consequently, it can be said that it is an important experimental center in the studies to be carried out to be minimally affected by earthquakes. Steel structures, which are more advantageous against earthquakes, can be designed with many different systems. In this study, selected a plot of land in the provincial center of Düzce, one of the cities located in the 1st degree risk zones in Turkey for examine the effects of design differences under horizontal and vertical loads. The moment transmitting frame and the centrally braced frame have been designed between the same axes in the same building design. This study was examined for 5 different two-dimensional models with brace and moment transmitting systems. Interstory drift analyses will be performed by comparing the reduced spectra according to the type of building structural system defined in the Turkey Building Earthquake Code (TBEC 2019) with the 1999 Düzce Earthquake Spectrum. Interstory drift analysis of all designs were completed and their advantages and disadvantages comparative were evaluated.
Today, cities, as dynamic, complex, and fragmented spaces, present various urban problems. Projects proposed through traditional planning approaches are inadequate in solving urban problems because they are resistant to change, have slow-moving processes, and are difficult and costly to remove when they fail to meet user needs. Therefore, tactical urbanism, one of the emerging new planning approaches, emerges as an approach that advocates site-specific and pinpointed methods, emphasizes low cost and low risk, and incorporates flexible and rapid interventions that adapt to change. This study aims to propose solutions for areas defined as "lost space" by utilizing these practices. The issue of reclaiming lost space, an urban problem to which tactical urbanism provides applications, has gained importance due to the density, congestion, and waste of urban land in today's cities. It is also clear that these areas pose problems such as pollution, neglect, urban disconnection, insecurity, and idleness. This study examines lost spaces through a comparative analysis of national and international examples using the tactical urbanism approach. The study determined that tactical urbanism practices are implemented globally at many different scales and functions, and that stakeholder diversity is equally high. The study concluded that tactical urbanism practices are an effective solution for lost spaces. Furthermore, the Kadıköy examples were found to be successful compared to their international counterparts due to their low cost and diversity in terms of participation. However, they were found to be inadequate in terms of functional diversity and flexible spatial production methods.
The growing energy demand and environmental concerns have increased the orientation towards energy- efficient and environmentally friendly technologies. In this context, transcritical heat pumps using natural refrigerant R744 (CO₂) stand out. However, real-time monitoring is critical for operating and managing these systems at optimum performance. This study aimed to develop a unique system to monitor and analyze the performance of a transcritical CO₂ heat pump using Internet of Things (IoT) technologies. The system is based on Raspberry Pi and Arduino microcontrollers, various sensors (temperature, pressure, flow, power), a MySQL database, and a dual interface (Lazarus-based local and PHP/JavaScript/HTML-based web). While Arduino collected and processed sensor data using C language, the object-oriented software developed with Lazarus on Raspberry Pi received, processed, stored the data, and presented it on the local interface. Simultaneously, the web interface provided remote access and visualization of the data. The developed system successfully monitored the critical performance parameters of the heat pump (temperature, pressure, flow, power consumption, data required for COP calculation). The results indicate that the proposed IoT-based monitoring system has the potential to increase the energy efficiency of transcritical heat pumps, facilitate operational management, and enable early detection of potential failures.
This study aims to evaluate and quantify wind turbine performance aloft using a field-deployable tethered aerostat test platform, combining a lightweight horizontal-axis turbine, a conical diffuser, and synchronized onboard measurements of atmospheric conditions and electrical output. The platform is designed to enable higher-altitude operation (up to ~500 m), while the results presented in this study are based on field measurements up to ~60 m. Field measurements show that the mean wind speed increased from approximately 4.6 m/s at 20 m to 6.1 m/s at 60 m, producing a corresponding rise in electrical power output from ~37 W at the lowest recorded operating point (V≈2.6 m/s) to a maximum of ~52 W at around 50 m altitude (V≈6.1 m/s), consistent with the cubic wind–power relationship. The conical diffuser was additionally assessed under comparable wind conditions (~6.1 m/s) and provided a modest gain from 50 W to 52 W (≈ 4%) for the present geometry. These results indicate that height-adjustable lighter-than-air deployment can improve energy yield in low-wind regions without tower infrastructure by leveraging the vertical wind gradient, while highlighting that diffuser benefits are measurable but limited under the tested conditions and merit further geometric optimization and repeated trials.
Tatvan (Kotum) Stream is a water source formed by small branches fed from springs around Nemrut Mountain. The watershed is located in the high and rugged terrain of the Eastern Anatolian Plateau, where basaltic and andesitic rocks are common. Due to high snow cover in winter and rapid snowmelt in spring, seasonal changes are observed in the stream regime. Hydrologically, Tatvan Stream is one of the important tributaries flowing into Lake Van. The alluvium carried by the stream increases the flow rate in spring and contributes to delta formation around the lake. However, in recent years, increasing settlement and agricultural activities have threatened the water quality. This study examines heavy metals and other chemical parameters in water samples from Kotum Stream within the Lake Van Basin. Heavy Metal Pollution Index (HPI) and Metal Index (MI) were calculated to assess pollution levels. Additionally, other water quality parameters were evaluated and interpreted according to regulatory standards. The aim of the study is to provide a scientific basis for developing sustainable water management policies and environmental protection strategies for the region.
This study aims to determine the most suitable locations for electric vehicle charging stations within the borders of Düzce province. A p-median-based Genetic Algorithm (GA) method was used in the site selection process. As an alternative solution approach, the Binary Particle Swarm Optimization (BPSO) algorithm was utilized. Detailed spatial data covering the Düzce city center and its surroundings were used in the study; 44 potential station points were identified, and 5,000 demand points to be directed to these points were defined. Different selection and mutation operators were tested within the GA method to determine the most suitable charging station locations. Operators such as Random Solution, Tournament Selection, and Roulette Wheel were compared. The study specifically examined which method provided the most efficient distribution for a region like Düzce. According to the results obtained, the Tournament Selection method yielded more successful results in terms of both cost and performance compared to other operators. Spatial analyses show that the model accurately reflects areas with high demand. Furthermore, it was observed that the most efficient solutions are clustered in specific areas. Another aim of the study is to comparatively evaluate the results obtained from the GA and BPSO methods. The findings revealed that BPSO offers faster and higher-quality solutions, especially in binary positioning problems. In this respect, BPSO stands out as a strong and feasible option for charging station planning. In conclusion, this study makes significant contributions to literature, both methodologically and practically. In analyses conducted with real field data, the GA and BPSO algorithms were compared via the p-median model; valuable information was obtained regarding the performance of these heuristic methods in complex urban structures.
Contemporary businesses must evaluate the performance of their sales personnel and refine their sales strategies. In this context, a variety of approaches are employed to develop strategies, including combining sellers based on their respective sales characteristics, to increase sales. Clustering, a machine learning approach, is used to derive inferences from sales data. The results are then used to inform future sales planning and determine priorities. To achieve this, the sellers are initially grouped (clustered) by similar characteristics based on specific criteria (such as sales volume and product information). This enables the identification of the typical strengths and weaknesses of sellers within each cluster. To illustrate, while sellers in a cluster with high sales volume and customer satisfaction scores may assume a pioneering role in the introduction of new products, it may be beneficial to investigate which products could be preferred in the region where sellers in a low-performing cluster are located, and what measures could be taken to increase sales of these products. By examining the sales performance of clustered sellers, it is possible to ascertain the relationships among the best-selling products across different applications. This approach enables the identification of products sold in conjunction, products that stimulate each other's sales, and products that appeal to disparate customer segments. Following the cluster analysis, an association analysis enables a more comprehensive investigation of the interrelationships among products. The results of this analysis permit the identification of product preferences among specific customer profiles. Based on the information mentioned above, more effective product recommendations and personalized marketing strategies can be formulated. An examination of sales within the identified clusters reveals pertinent information.
This study explores the use of sustainable natural resources in wastewater treatment, focusing on Paronychia carica, a member of the Magnoliopsida class, as a biosorbent for removing Procion Red dye (PR) from aqueous solutions. The effects of pH, initial dye concentration, adsorbent dosage and contact time on the adsorption performance were systematically investigated. Maximum dye uptake (83%) was achieved after 4 hours with an initial dye concentration of 30 mg L−1 and 0.5 g ofadsorbent at a mildly acidic pH of 4.0. Kinetic analysis revealed that the adsorption process followed the pseudo-second-order (PSO) model, with the calculated equilibrium capacity closely matching the experimental value, highlighting the dominant role of available adsorption sites. Equilibrum data were best described by the Freundlich, Dubinin-Radushkevich and Temkin isotherm models, confirming adsorption on a heterogeneous surface. The mean free adsorption energy from the D–R model (E = 8.53 kJ mol⁻¹) indicated that the physisorption is the primary mechanism. These findings demonstrate that P. carica can efficiently and rapidly remove PR dye under mild conditions. Overall, the study highlights the potential of P. carica as a low-cost, sustainable biosorbent for textile wastewater remediation, promoting eco-friendly treatment and supporting global efforts to reduce industrial pollution.
Hydrogen-enriched combustion is central to decarbonizing high-efficiency energy systems, yet its practical adoption is limited by the onset of thermoacoustic and hydrodynamic instabilities in premixed flames. In this study; a novel, integrated framework that combines high-fidelity computational fluid dynamics (CFD) simulations with interpretable deep learning for the prediction and physical diagnosis of combustion instability was proposed. A parametric suite of 1,500 axisymmetric CFD simulations was carried out, systematically varying hydrogen blending ratios (0–100% by volume), equivalence ratios (ϕ = 0.6–1.4), and turbulence intensities (5–25%). Key instability markers including root-mean-square (RMS) pressure, flame front wrinkling, and radical pool dynamics were extracted from both stable and unstable flame regimes. The data collected was used to train a hybrid convolutional neural network–long short-term memory (CNN–LSTM) model, which achieved a test accuracy of 94.3%, F1-score of 94.4%, and area under the receiver operating characteristic curve (AUC-ROC) of 0.978 in binary regime classification. SHAP-based interpretability analysis demonstrated that the model’s predictions were grounded in physically relevant features, with RMS pressure, OH fluctuations, and dominant acoustic frequencies serving as the principal contributors. AI-predicted instability regime maps showed an 88.6% overlap with CFD-derived instability thresholds, highlighting the physical consistency of the approach. Distinct field visualizations showed that unstable regimes (ϕ = 1.1, H₂ = 80%) exhibit pronounced front wrinkling, broader high-temperature zones, and spatially distributed radical production compared to stable flames. This approach opens a promising path for data-driven, physically interpretable instability diagnostics, which could directly impact for burner design, operational safety, and real-time combustion monitoring in hydrogen-based systems. In future work, it is aimed to extend this approach to multi-fuel configurations and experimental integration for real-world deployment.
Artificial Neural Networks (ANN) have gained popularity again due to the increasing interest and developments in artificial intelligence, as well as the increased computational power offered by High Performance Computing (HPC) systems. Since neural network applications are used in large data centers and HPC systems, they face similar reliability issues such as bit slippage in registers and memory structures that are common in these systems. Therefore, they require special robustness and protection mechanisms that can significantly increase the system cost. However, understanding the impact of hardware failures on different components of ANN applications can help determine which parts are more vulnerable and require higher reliability. In this study, the effects of hardware faults on ANN applications when they are run in HPC systems and large-scale data centers are evaluated, and thus, the reliability costs are aimed to be reduced. Fault injection experiments performed with traditional techniques can be quite time-consuming for ANN applications. Therefore, a method is presented to reduce the fault injection time in such applications. When we evaluate the effects of hardware faults on Artificial Neural Network (ANN) applications running on CPU-based (Intel Xeon) and GPU-based (NVIDIA V100) high-performance computing (HPC) systems, our results show that ANNs are vulnerable to some hardware faults, especially those occurring in certain layers and architectural registers.
This study discusses deformation problems in field sprayer arms encountered in the literature. It compares the static stress and fatigue analysis of the sprayer arm designed with St 37 and St 44 materials. The sprayer arm is fully open for St 37 and St 44 structural steels with the weight force in the vertical direction y-axis generated by its weight effect in the analyses. According to the analysis results, the highest principal stress values and total displacement values are the same for St 37 and St 44 materials; the highest principal stress value is 87,147 MPa, and the highest total displacement value is 7,1729 mm. The safety coefficient for static stress analysis was the highest value for St 44 material and was 3,1556. The highest fatigue life belonged to the St 44 material; the result was 2,287×10^6 cycles. The highest minimum safety factor also belongs to St 44 material, and its value is 2,9273.
In this study, new patch antenna designs operating in the WiMAX and ZigBee bands are proposed for smart grid data communication. The same design procedure has been repeated for different dielectric materials such as FR4, RO3203, RT6006, used in the patch antenna structure. This way, it has been determined which material exhibits more efficient propagation characteristics at the target frequency bands. Thanks to the proposed rectangular slots on the antenna, improvements have been achieved both in return losses and in operational bandwidths. The slot dimensions have been chosen as specific multiples of the wavelength for each material, enabling the generalization of the design procedure. Moreover, the width and length of the first slot were taken as λ/50 and λ/20, respectively, which resulted in an increase in the antenna’s bandwidth. Then, the dimensions of the second rectangular slot were defined relative to the first slot, with its width set to half of the first slot’s width and its length set to four times the first slot’s length, yielding a further enhancement in the bandwidth. The antennas designed in the simulation environment have been fabricated using the printed circuit board (PCB) etching method. It has been demonstrated that the antennas measured with a vector network analyzer operate at the designed frequency as transmitter and receiver with the help of a signal generator and a spectrum analyzer.
Radiation shielding is a fundamental strategy for protecting individuals and the environment from the harmful effects of radiation exposure. Over the past two decades, a wide array of shielding materials has been developed to address the specific requirements of medical, nuclear, and space applications. This study employs a bibliometric approach to systematically analyze the scientific literature concerning radiation shielding materials. A total of 6,515 documents published between 2000 and 2023 were retrieved from the Scopus database and examined. The findings indicate that the most frequently investigated materials include silica, tungsten, polymers, carbon nanotubes, and lead. Emerging research trends—particularly in nanotechnology and composite materials—have also received significant scholarly attention. Keyword and co-word network analyses reveal that research is primarily concentrated on the concepts of “radiation shielding” and “radiation protection,” with strong thematic links to terms such as “gamma rays” and “neutrons”. This study underscores the diversity of materials and innovative approaches being explored in the field. Furthermore, it provides a structured overview and a guiding framework for future research, particularly regarding the development of cost-effective and eco-friendly shielding materials.
Temporal changes occur in Land Use/Land Cover (LULC) due to natural and anthropogenic impacts. It is important to determine the changes in LULC, to monitor their effects on ecological processes, and to contribute to sustainable land management, environmental policies and planning studies. Ayvacık, which is rich in natural and cultural heritage values, where tourism activities are active in coastal areas, provides the main source of livelihood with agriculture, and is located at a critical point in terms of transportation, has been experiencing a tendency to change in recent years due to these factors. In this study, it was aimed to determine the change in Ayvacık District of Çanakkale Province between 1985-1997-2010 and 2023 in a multi-temporal manner by using Remote Sensing (RS) and Geographical Information Systems (GIS) techniques. Machine learning based Support Vector Machines (SVM) algorithm was used in the classification process through ArcGIS 10.8 programme. As a result, the temporal and spatial change of the district was determined and change matrix and change map were produced. In 38 years, a total of 44184.42 hectares of land in the study area has been changed, while 43452.45 hectares of land has maintained the same class characteristics.
This study examines capacity inadequacies at airports, a critical component of airline transportation, and their impact on passengers. Despite the advantages of air travel, such as time savings and convenience, increasing demand often leads to long queues and delays at airports. Congestion, particularly at key points such as security checks, check-in counters, baggage drop-off, and immigration controls, significantly impacts passenger satisfaction. The study focuses on enhancing operational efficiency and reducing costs by minimizing waiting times and optimizing resource utilization in airport processing units. To evaluate these issues, a discrete-event simulation model was developed using domestic and international passenger flow data from İzmir Adnan Menderes Airport. The analysis is guided by the hypothesis that improvements in staff allocation and process configurations can reduce waiting times without substantially increasing operational costs. Simulation analyses, supported by insights from the literature, explore various scenarios for optimizing staff shifts and implementing modern solutions, such as online check-in. The findings reveal that reducing staff in specific processing units can lower costs but may increase waiting times, highlighting the trade-off between efficiency and service quality. This study provides valuable insights for airport managers to make strategic decisions and serves as a crucial reference for future operational planning.
Cancer is one of the most common and deadly diseases worldwide, and the development of new therapeutic strategies to overcome treatment resistance and improve clinical outcomes remains a critical challenge. Heterocyclic compounds have long been recognized as fundamental building blocks in medicinal chemistry. Among them, the isoxazole ring has emerged as a promising pharmacophore in anticancer drug design due to its unique electronic properties and hydrogen-bonding capability. Isoxazole is a five-membered heterocyclic ring containing nitrogen and oxygen atoms at the 1st and 2nd positions, enabling diverse interactions with enzymes and receptors and resulting in a wide range of biological activities. In this review, scientific studies on the development of isoxazole-containing compounds as anticancer agents have been systematically examined from the past to the present. The selected studies were evaluated based on their reported anticancer activity, molecular mechanisms of action, and clinical development status. Isoxazole derivatives have demonstrated significant activity against various cancer types, including breast, lung, colorectal cancers, and leukemia, through mechanisms such as apoptosis induction, cell cycle arrest, kinase inhibition, and suppression of angiogenesis. Moreover, the clinical approval of certain isoxazole-based drugs further supports the translational potential of this scaffold. Overall, the isoxazole core represents a promising and clinically relevant pharmacophore for the development of next-generation targeted anticancer therapies.
