SDG 9 · Sanayi, Yenilikçilik ve Altyapı · SDG · Erzincan Binali Yıldırım Üniversitesi

Yazarlar

Yayınlar

Yeniden eskiye Eskiden yeniye

ArticleAçık Erişim2023/r/ebyu/13461
A novel capacitor-voltage reduced bidirectional PWM DC-DC buck-boost converter for renewable energy battery charge system
Sahin, Yakup; Ting, Naim Suleyman; Yesilyurt, Huseyin
A novel capacitor-voltage reduced bidirectional (CVRB) PWM DC-DC buck-boost converter is presented in this study. Compared to the conventional bidirectional buck-boost converter, the proposed converter has a lower voltage rating filter capacitor. Accordingly, the given converter has a lower cost and higher power density than the conventional buck-boost converter. Additionally, the proposed converter is more efficient due to the direct power transfer feature. Besides, the semiconductor switches have no extra voltage/current stress. The theoretical analysis of the converter is made, and its mathematical analysis is presented. The novel converter is experimentally operated in both the buck and boost modes. The experimental waveforms are shown for both operations. The proposed converter is operated in 100 W output power and 20 kHz switching frequency conditions. The efficiency of the proposed converter is 95.6% for the boost mode operation and 95.1% for the buck mode operation at the nominal output power.
RDL Kalıcı Kaydıİncele
Book ChapterYalnız künye2023/r/ebyu/13384
The role of social manufacturing for entrepreneurship in the industry 4.0 transformation
Aka, Salih
Social manufacturing is an internet-and social network-based platform where different manufacturers' capabilities can be seen and manufacturers with similar skills and hardware levels can come together to form communities. In the horizontal community, manufacturers who have similar socialised manufacturing resources and can perform the same processing activities are brought together. In the vertical community, there are manufacturers who can take part in different production activities of a product, from raw material to final product, and combine their socialised resources in this direction. Social manufacturing allows the bargaining power of the main producer to decrease thanks to the communities it establishes. The social manufacturing platform is a virtual showcase for small-scale entrepreneurs to present their talents, an effective cost management mechanism for customers, and an interaction, coordination, and sharing tool that will maximise production efficiency for countries.
RDL Kalıcı Kaydıİncele
Book ChapterYalnız künye2023/r/ebyu/13312
Which barriers need more attention in food waste management and blockchain integration?
Aka, Salih
Blockchain is a digital innovative technology that is decentralized, open to the use of all stakeholders, does not allow data manipulation, and has many more features. Blockchain, which has the potential to solve problems in food safety, is attracting more and more attention from researchers in this field. In this study, a conceptual model for the integration of blockchain and food waste management is proposed. The combination of blockchain and food waste management has economic, social, and environmental dimensions that will serve the purpose of reducing food waste. However, some blockchain barriers make it difficult to integrate with food waste management. Blockchain barriers that affect the success of this integration are examined with the BEST-Worst Method (BWM). As a result of the analysis, infrastructure setup and adaptation challenges stand out as the most important barrier. The study would be a starting point for practitioners working on food waste management and the blockchain system, and government officials planning regulations in this area.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/13254
Experimental and computational study on local buckling of standard and improved cold-formed square hollow sections under static and dynamic loading
Cosgun, S.I.
Because of aesthetics, economic advantages, easy composite member production and design, and a wide range of standard sizes, hollow sections are commonly utilized in various construction industry applications worldwide, including conventional and unique free-form structures. However, the sections must fulfill more restricted circumstances to avoid early local buckling. The behavior of cold-formed square hollow sections with different wall thicknesses under static and dynamic transverse impact loading was investigated experimentally and numerically in this study. A method that uses rivet nuts is proposed to improve the dominant local buckling behavior, which is observed in the experimental studies and causes a relatively low bending strength, for providing convenience in field applications. The effectiveness of the proposed method is demonstrated experimentally and numerically using finite-element analysis. In addition to conventional measurement techniques, motion analysis is performed using digital image correlation to measure the displacements during static and dynamic tests. With the proposed technique, the local buckling strength of the square hollow sections is increased by 64% under static loading and by approximately 40–85% under dynamic loading compared with standard square hollow sections, respectively. This is due to the internal stiffness plate, steel threaded rods, and rivet nut configuration connecting the flanges and webs of the square hollow section, which ensures load sharing between the section parts. Additionally, steel threaded rods prevent early buckling of the internal stiffness plate. In addition, parametric analysis results show that the thickness of the internal stiffness plate has a negligible effect on the local buckling strength, although the presence of an internal stiffness plate increased the strength. The numerical analysis results indicate that the local buckling behavior of both conventional and improved square hollow specimens can be predicted with acceptable errors in terms of their strength and damage modes using finite-element analysis models.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/13240
The relationship between FDI, CO2 emissions, and energy consumption in Asia-Pacific economic cooperation countries
Balli, Esra; Sigeze, Ciler; Ugur, Mehmet Sedat; Çatık, Abdurrahman Nazif
This paper investigates the relationship between CO2 emissions, energy consumption, economic growth, and foreign direct investment for a sample of Asia-Pacific Economic Cooperation Countries (APEC) countries from 1981:Q1 to 2021:Q1 employing panel data methodology. We identify cross-sectional dependence and hence utilize the cross-sectional augmented Dickey-Fuller panel unit root test for appropriate estimation. The cointegration test developed by Westerlund (2008) reveals a long-run equilibrium between CO2 emissions, energy consumption, economic growth, and foreign direct investment. Long-run parameter estimates based on Common Correlated Effect Mean Group indicate that an increase in FDI inflows has a negative impact on air quality, supporting the pollution haven hypothesis. The cointegration test results also show that the impact of Gross Domestic Product (GDP) on CO2 emissions varies by country in the estimation sample. In contrast to the mixed evidence on the effects of other variables, the increase in energy consumption is positively and significantly affecting CO2 emissions in all APEC countries. Emirmahmutoglu and Kose Econ Model 28:870-876, (2011)’s panel causality test results show a bidirectional relationship between FDI and CO2 emissions in Japan. Furthermore, there is a bidirectional causal relationship between GDP and energy consumption in Australia, China, Japan, and Singapore. Overall, empirical evidence suggests that APEC countries should adhere to strict regulations and invest in environmental-friendly clean technologies to attract foreign direct investment.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/13216
Computational modeling of the axial behavior of corroded and buckled short steel piles strengthened using concrete-filled GFRP jackets
Cosgun, S.I.; Kaya, A.; Gencturk, B.
Steel columns or piles commonly used in bridges and marine structures in various parts of the world exhibit insufficient strength due to increased load demands or section loss due to corrosion. The latter issue can cause unexpected buckling of these columns. In this study, the repair of corroded and buckled short steel piles using concrete-filled GFRP jackets was numerically simulated to better understand the overall effectiveness of the method and the influence of the parameters involved. Thirteen steel piles used in this study had different degradation characteristics to investigate the section losses affecting the axial capacity and failure modes. The parameters considered included the presence or absence of concrete steel reinforcement (rebar), the diameter of the rebar, the number of GFRP layers, the location of the worn zone in the steel section and the severity of the section loss. The numerical models were validated using these experimental results of the repaired specimens from the authors’ earlier work on the topic. The main contribution of this study is in the development of the numerical models that can very accurately simulate the ultimate strength and damage behavior of retrofitted piles that are heavily corroded and buckled. The validated models were used in a parametric study to understand the effects of different levels of concrete strength, or corrosion of the piles in the flanges, the web, or both. The results of the parametric study provided further understanding of the structural behavior of degraded and strengthened steel piles.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/13215
Prediction of flood routing results in the Central Anatolian region of Türkiye with various machine learning models
Katipoğlu, Okan Mert; Sarıgöl, Metin
Flood routing models are vital in predicting floods and taking all necessary precautions in the region where floods occur, preventing loss of life and property in the region and protecting agricultural areas. This study aims to compare the performance of various machine learning models such as Bagged Tree, Gradient-Boosted Machine, Random Forest, K-Nearest Neighbor, Support Vector Machine and Extreme Gradient Boosting for flood routing prediction models in Ankara, Eskişehir and Sivas. In addition, the predictive success of tree-based algorithms established according to the optimized and default parameters was compared. For this purpose, the flood data of 2013, 2014 and 2015 discharge observation stations located in Ankara D12A242-D12A126, D12A170-D12A172 in Eskişehir and D15A290-E15A035 in Sivas were used. While establishing the machine learning (ML) models, the data was selected as 80% training and 20% testing. Model performances were tested according to various statistical indicators such as root mean square error, mean absolute error and determination coefficient. As a result of the study, the Gradient-Boosted Machine was chosen as the most successful model in estimating flood routing. In addition, the K-nearest neighbor model with 3-nearest neighbor achieved high-level prediction success with the lowest error rates in Ankara. The findings are important in terms of flood management and taking necessary precautions before the flood occurs.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/13200
The Effect of Virtual Inertia and Damping Control on the Stability Region of Load Frequency Control Systems with Time Delays
Hasen, Suud; Aydın, Ömer; Ayasun, Saffet; Sönmez, Şahin
This article examines the effect of virtual inertia and damping (VID) control on the stability regions of time-delayed load frequency control (LFC) systems. Because of the substantial integration of renewable energy sources (RES) and the utilization of an open communication network for sending/receiving data and control signals, the system inertia significantly decreases, and inevitable delays are observed, both of which adversely affect the controller performance and frequency stability. For improving the stability of LFC systems including RES and delays, this study proposes the integration of the VID control and determines stability regions using the method of stability boundary locus. The regions’ boundaries are substantiated by time-domain simulations and a quasi-polynomial mapping-based root finder (QPmR) algorithm. A wide-ranging evaluation of the VID coefficients’ effects on the stability region and the system’s frequency response are presented. Results evidently divulge that with the integration of the VID control, stability regions get larger, the frequency dynamic is improved, and unstable LFC systems with delays are stabilized.
RDL Kalıcı Kaydıİncele
Conference ObjectYalnız künye2023/r/ebyu/13033
Design of an Optimal PI Controller for Time-Delayed Two-Area LFC-VID System
Hasen, Suud Ademnur; Aydin, Omer; Ayasun, Saffet
Growing deployment of renewable energy sources (RES) may lead to oscillations in the system frequency caused by reduced inertia, while communication networks generate time delays that might jeopardize system dynamic performance and induce additional frequency fluctuations. This paper presents the optimization of Proportional Integral (PI) controller gains in a time-delayed two-area load frequency control (LFC) system with virtual inertia and damping (VID). Optimization techniques such as Crow Search Algorithm (CSA), Particle Swarm Optimization (PSO), Vortex Search Algorithm (VSA), and Sine-Cosine Algorithm (SCA) are used to optimize the PI controller gains to obtain best frequency performance of the system. In all algorithms, the integral square error (ISE) objective function is employed during the optimization process. Results are confirmed via time-domain simulations in Matlab/Simulink.
RDL Kalıcı Kaydıİncele
ArticleAçık Erişim2023/r/ebyu/13011
Determining the regional disaster risk analysis of buildings in Erzincan; [Određivanje regionalne analize rizika od katastrofa za zgrade u Erzincanu]
Başgöze, Ayşe; Güncü, Arzu
Erzincan is located in a region sensitive to seismic activities, which can destroy buildings and lead to loss of life because of severe earthquakes. Furthermore, analysing and determining structural risks is important because it reduces possible disaster risks and provides mitigation strategies. This study aims to assess the seismic vulnerability of existing buildings in Erzincan using a rapid visual screening method and create a regional-scale inventory. Furthermore, 490 residential reinforced concrete buildings in five neighbourhoods were analysed using a street scanning method (first-level evaluation) developed by METU, and maps were created using the ArcGIS program. The results revealed that poor construction quality, soft ground, and heavy overhang are the main vulnerability parameters that change the risk priority range of reinforced concrete residential buildings. Conversely, poor construction quality affected most first-priority buildings. Therefore, there is a need for effective seismic mitigation planning for Erzincan, as 49 % of buildings in the surveyed neighbourhoods required a second-stage assessment. In addition, the method ranks the buildings according to their risk priorities, and the obtained data on the map provided useful information for effective strategies for implementing risk reduction policies in Erzincan.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/13010
Design and Implementation of a Flyback-Assisted Wide-Range Fully Soft-Switched T-Type Inverter
Ting, Naim Suleyman; Sahin, Yakup; Ozgenel, Mehmet Cihat
A novel soft-switched three-level T-type inverter is proposed in this article. The proposed inverter provides turning-on by zero-voltage transition and turning-off by zero-voltage switching (ZVS) for main switches without any voltage or current stresses. The switching energies are transferred to the snubber capacitors, so the turning-off process of the main switches is improved by these snubber capacitors. Besides, the auxiliary switches are turned on by zero-current switching and turned off by ZVS. Furthermore, all snubber diodes are operated by soft-switching (SS), and the proposed SS snubber cell decreases the electromagnetic interference noises. The novel snubber cell has a simple structure, ease of application, common-ground switches, and low-cost features. The theoretical analysis of the inverter is clarified, and the operating modes for steady-state analysis are presented in detail. The experimental results rated 1-kW output power and 100-kHz switching frequency are provided to justify the theoretical analysis. The proposed unique snubber cell can be easily implemented in other three-level inverters, and the practice is demonstrated in the article.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/12989
Coupling machine learning with signal process techniques and particle swarm optimization for forecasting flood routing calculations in the Eastern Black Sea Basin, Türkiye
Katipoğlu, Okan Mert; Sarıgöl, Metin
With the effect of global warming, the frequency of floods, one of the most important natural disasters, increases, and this increases the damage it causes to people and the environment. Flood routing models play an important role in predicting floods so that all necessary precautions are taken before floods reach the region, loss of life and property in the region is prevented, and agricultural lands are protected. This research aims to compare the performance of hybrid machine learning models such as least-squares support vector machine technique hybridized with particle swarm optimization, empirical mode decomposition, variational mode decomposition, and discrete wavelet transform processes for flood routing estimation models in Ordu, Eastern Black Sea Basin, Türkiye. In addition, it is aimed to examine the effect of data division in flood forecasting. Accordingly, 70%, 80%, and 90% of the data were used for training, respectively. For this purpose, the flood data of 2009 and 2013 in Ordu were used. The performance of the established models was evaluated with the help of statistical indicators such as mean bias error, mean absolute percentage error, determination coefficient, Nash–Sutcliffe efficiency, Taylor Diagrams, and boxplot. As a result of the study, the particle swarm optimization least-squares support vector machine technique was chosen as the most successful model in predicting flood routing results. In addition, the optimum data partition ratio was found to be Train:70:Test:30 in the flood routing calculation. The findings are essential regarding flood management and taking necessary precautions before the flood occurs.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/12972
Mechanical, durability and microstructural characteristics of Portland pozzolan cement (PPC) produced with high volume pumice: Green, cleaner and sustainable cement development
Ulusu, Hayati; Yılmaz Aruntaş, Hüseyin; Burcu Gültekin, Arzuhan; Dayı, Mustafa; Çavuş, Murat; Kaplan, Gokhan
Nowadays, it is frequently preferred to use pozzolans instead of clinker to reduce the CO2 emissions that increase with cement production. Because in the future, the release of artificial pozzolans, such as fly ash and blast furnace slag, will decrease. Therefore, the demand for natural pozzolanic cement will increase. This study aimed to produce Portland Pozzolan Cement (PPC) by using pumice instead of clinker up to 40%. The Chemical, physical, mechanical, durability, microstructure, and carbon footprint characteristics of PPCs were investigated experimentally. As the pumice volume in the PPCs increased, the CaO/SiO2 ratio and specific gravity decreased. Due to the hygroscopic moisture of the pumice, its grindability has become difficult. While the 40 µm sieve residue of PPCs ranges from 0.8 to 1.7%, the sieve residue ratio generally increases as the pumice volume increases. The porous structure of pumice led to an increase in Blaine fineness as the replacement ratio increased. The Blaine fineness of PPCs containing 40% pumice rose by about 20% more than the reference cement. The consistency water of PCs varies between 29 and 36%. As the water consistency increased, the setting time of the pastes was prolonged. Initial setting times of PPCs are between 150 and 220 min. If the CaO/SiO2 ratio decreased with the pumice volume, the compression strength of the mortars decreased. While the 28-day compression strength of PPC containing 10% pumice was measured as 36.2 MPa, this value was measured as 21.6 MPa for PPC containing 40% pumice. As the pumice volume increased, the mortars' drying shrinkage and sulfate-induced expansion values decreased significantly. According to the sustainability analysis, the CO2 emissions of cement vary between 518 and 880 kg CO2/t. By using pumice instead of clinker, CO2 emissions of cement can be reduced. In SEM examinations, it was determined that pumice concentrated the microstructure. As a result, it has been determined that eco-friendly cement can be obtained by using 10% and 20% pumice in PPC production.
RDL Kalıcı Kaydıİncele
Conference ObjectYalnız künye2023/r/ebyu/12947
Heuristic Optimization of PI Controllers for Time-Delayed LFC-EV Systems
Aydin, Ömer; Hasen, Suud Ademnur; Ayasun, Saffet
This study proposes an optimization method for the proportional-integral (PI) controller gains of a single-area time-delayed load frequency control system with electric vehicles (LFC-EV). Four heuristic algorithms are used to optimize the PI controller gains: crow search algorithm (CSA), particle swarm optimization (PSO), vortex search algorithm (VSA), and sine-cosine algorithm (SCA). The integral square error (ISE) objective function is used in the optimization process for all algorithms. Finally, a time-domain simulation of the LFC-EV system is conducted to evaluate the performance of the proposed algorithms. The optimized PI controller gains exhibited superior performance compared to randomly selected gains, leading to decreased oscillations and faster settling times, as evidenced by the results.
RDL Kalıcı Kaydıİncele
ArticleAçık Erişim2023/r/ebyu/12935
Investigation of the effect of the gap at the zero-crossing point of PWMs creating the first level voltage in a multi-level inverter
Can, Erol; Sayan, Hasan Hüseyin
The usage of multi-level inverters in converting direct current to alternating current is one of the most demanded and used methods. While creating multi-level voltage with inverters, the operating times of the switches that constitute the first step of the alternating voltage determine the gap of the step at the zero-crossing point. This situation can be preferred to create an extra level. Therefore, this article examines the effect of variation in operating times and frequencies of first-step PWMs on this zero-crossing point. First, the inverter structure providing load voltage is given in the study. Then, the transition of PWM run times is tried for different first-level voltages on the loads. The results obtained show the operation times of PWM at the transition point are quite effective on the load currents and voltages.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/12928
A novel conflict free routing with multi pickup delivery tasks for autonomous vehicles
Sarıçiçek, İnci; Yazıcı, Ahmet; Aslan, Özge
Purpose: This study aims to propose a novel method for the conflict detection and eradication of autonomous vehicles which has predetermined routes to establish multi pickup and delivery tasks according to task priorities and vehicle capacity status on each pickup and delivery nodes in assembly cells in the automotive production. Design/methodology/approach: In the designed system, the routing of autonomous vehicles (AVs) and scheduling of pickup and delivery tasks are established in production logistics. Gantt chart is created according to vehicle routes, and conflicts are detected using the proposed conflict-sweep algorithm. The proposed conflict-solving algorithm eliminates conflicts on intersections and roads by considering vehicle routes and task priorities. Findings: In many production systems, there is a need to obtain flexible routes in each pickup delivery task group that changes during day, week, etc. Proposed system provides remarkable advantages in obtaining conflict-free routes for pre-scheduled multi transport tasks of vehicles by considering efficiency in production systems. Originality/value: A novel method is proposed for the conflict detection and eradication of AVs. Proposed system eliminates conflicts on intersections and roads by considering pre-planned vehicle routes for a fleet of heterogeneous AVs. Unlike most of the other conflict-free algorithms, in which conflicts are solved between two points, proposed system also considers multi pickup and delivery points for AVs. This is pioneering paper that addresses conflict-free route planning with backhauls and scheduling of multi pickup and delivery tasks for AVs.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/12795
Improving the accuracy of rainfall-runoff relationship estimation using signal processing techniques, bio-inspired swarm intelligence and artificial intelligence algorithms
Katipoğlu, Okan Mert; Sarıgöl, Metin
In water resources engineering, streamflow estimation models are of great importance. The use of black box models in determining streamflow estimation is preferred because it saves time compared to deterministic models. In addition, the data needed is less in quality and quantity and has a lower transaction volume. Therefore, accurate flow estimation plays a crucial role in water resources planning, management, and sizing of water structures. This study aimed to forecast the monthly streamflow using particle swarm optimization, least-squares support-vector machine, and signal processing models. For this purpose, 575 monthly data between 1963-2010 were used in Sivas, and 622 monthly data between 1960-2011 were used in Kayseri. The model structure used past streamflow, temperature, and precipitation values as input. During the installation of hybrid models, the inputs are divided into subcomponents with three levels of discrete Meyer mother wavelet and presented to the LS-SVM model. The performance of the models was evaluated with the help of statistical indicators such as determination coefficient, mean absolute error, Nash–Sutcliffe Efficiency, mean square error, Violin plots and visual Taylor diagrams. At the end of the study, EMD-LSSVM was the best streamflow estimation model at station 1501, and W-LSSVM was the most successful streamflow estimation model at station 1535.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/12690
Fully soft-switched half-bridge DC–DC converter for hybrid electric vehicle battery charge applications
Yesilyurt, Huseyin; Sahin, Yakup; Ting, Naim Suleyman
This paper proposes a soft-switching (SS) active snubber cell for half-bridge (HB) DC–DC converters, which are utilized in hybrid vehicle battery charge applications. The proposed converter has an advantage over conventional HB DC–DC converters, which suffer from transformer leakage inductance issues. The new snubber cell performs SS operation using the energy stored in the leakage inductance. Thus, the leakage inductance, which creates troubles for HS DC–DC converters, is beneficial to the proposed converter. The proposed SS HB DC–DC converter is compared with the conventional HB DC–DC converter under 100 kHz and 110 W output power operating conditions. The efficiency of the proposed converter is 7.3% and 4.8% higher than the conventional HB converter at light and heavy loads, respectively. The proposed active snubber cell provides complete SS for the main and auxiliary semiconductor switches. The new snubber cell has a basic structure, low-cost, ease of application, and simple control features.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/12627
Experimental and statistical investigation of mechanical properties and surface roughness in additive manufacturing with selective laser melting of AlSi10Mg alloy
Siyambaş, Yusuf; Turgut, Yakup
In metal-based additive manufacturing, laser power, scanning speed, hatching distance and layer thickness are important manufacturing parameters. These parameters are very important in terms of microstructure, mechanical properties and surface quality of the final product. For this reason, it is very important choosing these parameters correctly in the production process. One of the common approaches used to optimize these parameters is volumetric energy density. In this study, product quality improvement in additive manufacturing of AlSi10Mg alloy was focused, and test pieces were produced at different volumetric energy densities. The parts were manufactured according to the Taguchi L9 orthogonal array experimental design. Three different laser powers (250–300–350 W), three different scanning speeds (900–1000–1100 mm/s), three different hatching distances (0.17–0.19–0.21 mm) and constant layer thickness values were used as manufacturing parameters. The performance properties of the manufactured parts were investigated in terms of mechanical properties and surface roughness. As a result of the study, it was observed that the mechanical properties of the samples produced at very high and very low volumetric energy densities decreased and the surface roughness increased. Maximum mechanical properties within the produced samples were obtained at 55.82 J/mm3 volumetric energy density, and minimum surface roughness was obtained at 54.46 J/mm3 volumetric energy density. For mechanical properties and surface roughness, multiple response optimization was performed with Taguchi-based Grey relational analysis method. According to the experimental results, optimum manufacturing parameters for additive manufacturing of AlSi10Mg alloy; it was obtained as 350 W laser power, 1100 mm/s scanning speed, 0.19 mm hatching distance and 55.82 J/mm3 volumetric energy density, which is the interaction of these parameters. The mechanical properties and surface roughness obtained graphically were supported by fracture surface images and microstructure studies. With multiple response optimization, 30.15% improvement was achieved in performance indicators. In addition, it has been demonstrated that volumetric energy density can be used as an important evaluation tool in determining product quality.
RDL Kalıcı Kaydıİncele
ArticleYalnız künye2023/r/ebyu/12617
Estimation of hourly flood hydrograph from daily flows using machine learning techniques in the Büyük Menderes River
Sarıgöl, Metin; Katipoğlu, Okan Mert
Flood routing calculations are of vital importance in estimating the floods occurring in the downstream region and taking all necessary measures to minimize the damages that the flood may cause. The prediction success of the created algorithms in the flood routing analysis was tried to be measured by training the daily flood data with machine learning algorithms. This research trained the daily average flood hydrograph data with regression tree, Gaussian process regression, support vector machine, ANFIS and regression tree ensembles. Then the flood routing calculations were made using these models for the hourly flood hydrograph of 2015. The success of the created machine learning algorithms was compared with statistical criteria such as mean absolute error, the root of mean squares of errors and coefficient of determination (R 2), and visually with Taylor diagrams and boxplots. As a result of the analysis established, the ANFIS model divided the inputs into 5 subsets, and the trim membership function showed good results in flood routing. It has been proved that with limited flow data, floods can be postponed, and loss of life and property can be prevented.
RDL Kalıcı Kaydıİncele