DSpace@ATÜ (Adana Alparslan Türkeş Bilim ve Teknoloji Universiti)
Not a member yet
3463 research outputs found
Sort by
Monthly streamflow prediction and performance comparison of machine learning and deep learning methods
No full textStreamflow prediction is an important matter for the water resources management and the design of hydraulic structures that can be built on rivers. Recently, it has become a widely studied research field where data obtained from stream gauge stations can be utilized for creating estimating models by resorting to different methods such as machine and deep learning techniques. In this study, we performed monthly streamflow predictions by using the following data-driven methods of machine learning: linear regression, support vector regression, random forest and deep learning (DL) models to compare the performances of ML's and DL's techniques. A general workflow that can be applied to similar regions is presented. An estimating model containing six-input combinations and time-lagged streamflow data is improved by means of the autocorrelation function (ACF) and partial autocorrelation function (PACF). Furthermore, moving average is used as a smoothing technique to make the dataset more stable and reduce the effects of noise data. A comparative evaluation has been conducted to determine the performances of the above-mentioned methods. In this study, we proposed four different DL models and compared them with existing techniques. For the comparison of the results, we used evaluation criteria such as Nash-Sutcliffe efficiency (NSE), mean square error (MSE) and percent bias (PBIAS). The experimental results indicate that our bidirectional gated recurrent units (BiGRU) model outperforms both ML algorithms and existing solutions with 0.971 NSE, 0.001 MSE and - 1.536 PBIAS scores
An Isolated High-Power Bidirectional Five-Level NPC Dual Active Bridge DC-DC Converter with Anti-Windup PI Controller for Electric Vehicle-to-Home Application
No full textOrange TM; STC15th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2023 -- 29 June 2023 through 30 June 2023 -- Bucharest -- 191300This paper introduces the design and analysis of an isolated bidirectional five-level (5L) neutral point clamped (NPC) dual active bridge DC-DC converter with an improved anti-windup strategy using proportional-integral (PI) controller for electric vehicle-to-home applications. The proposed converter is more advantageous in that it provides a 5L output voltage at the high-frequency isolation transformer sides thanks to its NPC structure compared to the conventional full-bridge circuit. Reducing switching stresses, providing galvanic isolation, ensuring bidirectional power flow and simple control strategy are the highlights of the proposed structure. A single phase shift (SPS) modulation strategy has been employed to regulate the output voltage and to generate switching signals of the proposed DC-DC converters' switches. To improve the dc-link voltage and to reduce the output voltage tracking errors, the anti-windup PI (PIAW) controller has been utilized instead of the conventional PI controller. To demonstrate the verification of the proposed converter and its controller, a simulation model has been developed in MATLAB/Simulink software. The performance and effectiveness of the proposed converter have been evaluated under steady-state and dynamic variations by the simulation results. © 2023 IEEE
The Numerical Simulation of Disturbed Region Corbels Containing Sustainable Concrete
No full textThe concept of sustainable concrete is aimed at using alternative materials instead of natural resources in concrete production. In recent years, many experimental and numerical studies have been carried out to show that sustainable concrete is an alternative to conventional concrete. In this study, a detailed numerical simulation of reinforced concrete corbels produced with sustainable concrete has been conducted comprehensively. According to the author's best knowledge, the novelty of this study: (1) this is the first time that comprehensive numerical simulation results of sustainable concrete corbels are reported in detail, (2) investigation of the applicability of existing numerical models in sustainable concrete corbels, (3) contribution the realistic simulation of the structural behavior of sustainable concrete corbels. It should also be highlighted that the variables included in the study were the recycled concrete aggregate ratio, the shear span-to-effective depth ratio, the tapering ratio, the angle between strut and tie element, and the amount of reinforcement. Furthermore, the numerical simulation results obtained by the finite element and truss analogy methods were extensively compared with the experimental results. In addition, a structural optimization design was executed on a short corbel to check the suitability of the selected truss analogy model. Based on the study, the simulation results obtained from the finite element method agreed fairly well with the experimental results. Furthermore, the truss analogy method provided relatively accurate and conservative results compared with the experimental results depending on the span-to-depth ratio. Besides, some recommendations for designing sustainable concrete corbels were also discussed
Comparison of Fuzzy Solution Approaches for a Bilevel Linear Programming Problem
No full text15th International Conference on Application of Fuzzy Systems, Soft Computing and Artificial Intelligence Tools, ICAFS 2022 -- 26 August 2022 through 27 August 2022 -- Budva -- 291409In this study, we consider solution approaches used to solve the proposed bilevel linear programming model for an Industrial Symbiosis network. We first solve this model with the well-known Karush-Kuch-Tucker (KKT) approach. However, transforming a bilevel programming model with the KKT approach increases the number of variables and constraints. For this reason, we use the fuzzy programming approach and the fuzzy goal programming approaches as alternatives to the KKT approach. Next, we compare the results of the KKT approach with these methods and examine the suitability of these approaches to solve our bilevel problem. Unlike previous studies, which claimed that fuzzy approaches mostly outperform the KKT approach, in our case, the best solution is obtained with the KKT approach. This is most probably because these approaches ignore the hierarchical nature of the problem. We believe that more research on fuzzy approaches is needed to evaluate the suitability of these approaches for solving bilevel programming problems. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.Türkiye Bilimsel ve Teknolojik Araştırma Kurumu, TÜBİTAK, (122M223
Promoting Migration-Oriented Resilience and Social Cohesion in Adana
No full textThis study analyses the vulnerabilities, opportunities and urban resilience and social cohesion awareness that emerged in Adana province in the face of the mass migration movement towards Adana province after the civil war that started in Syria in 2011. In this respect, we will first try to make conceptual fixations on resilience awareness. After explaining the historical origin and evolution of the concept, we will address the issue of putting the issue on the agenda of our country and municipalities at the local level. Afterwards, we plan to make use of the Migration Master Plan prepared by Adana Metropolitan Municipality with the perspective of increasing urban resilience against migration and establishing resilience. Based on these points, we would like to summarize the study under three main headings. In the introduction section, we will try to reveal the political and historical background of the issue at the scale of Adana. In the first section, we will discuss the issue of urban resilience in the face of migration by making conceptual fixations. In the second section, we will try to understand the impact of Syrian migration on the city in Adana. In the third part, we will discuss the vulnerabilities and opportunities set out in the Migration Master Plan prepared by Adana Metropolitan Municipality and discuss the problems, targets and solution vision. The study will conclude with a general evaluation on the subject
Türkiye’deki Belediyelerin Sorumluluk Sahasındaki Yollar için bir Derzli Donatısız Beton Yol Dizayn Kataloğu ve Modeli Önerisi
No full textBeton yollar yapısal ömürlerinin fazla oluşu, ekonomik, dayanıklı ve çevre dostu oluşları, her mevsimde uygulanabilir oluşları ve gece görüşünü olumlu yönde etkilemeleri gibi avantajları olan ve ülkemizde yerli imkanlarla üretimi gerçekleşen çimentonun en pahalı bileşeni olduğu ve dolayısıyla maliyet açısından da ön plana çıkan bir yol üstyapısı tipidir. Belediyelerin sorumluluk sahasındaki düşük hacimli yollarda kullanılabilecek Derzli Donatısız Beton Yollar (JPCP) için bir dizayn kataloğu ve modeli geliştirilmesinin birçok pratik faydası olacaktır. Bu çalışmanın amacı ülkemizde belediyelerin sorumluluk sahasındaki düşük hacimli JPCP yolların (hem sokak ve cadde gibi yollar (residential roads) hem de bulvar gibi daha yüksek hacimli yolların (collector roads)) tasarımında kullanılmak üzere: (1) farklı tasarım senaryoları ve parametreleri için PavementDesigner web programı kullanılarak beton yol plaka kalınlıklarını belirleyerek bir “Beton Yol Dizayn Kataloğu” oluşturmak ve (2) bu katalog kullanılarak çoklu regresyon modelleri geliştirerek alternatif senaryolar için hızlı bir beton plaka kalınlığı tahmin modeli geliştirmektir. Çalışmada 432 tasarım senaryosu belirlenmiş ve herbir senaryo için ilgili tasarım değerleri PavementDesigner’a girilerek beton plaka kalınlıkları elde edilmiştir. Farklı dizayn parametrelerinin beton plaka kalınlıklarına etkileri detaylı olarak sunulmuştur. Bununla birlikte bu 432 tasarım senaryosu kullanılarak hem sokak ve cadde gibi yollar (residential roads) hem de bulvar gibi daha yüksek hacimli yolların (collector roads) kalınlık tasarımında kullanılabilecek ve PavementDesigner çıktılarına başarılı bir şekilde benzer sonuçlar veren iki çoklu regresyon modeli geliştirilmiştir. Bu modeller kullanılarak belediyelerin sorumluluk sahasındaki yollar JPCP olarak hızlı bir şekilde tasarlanıp vakit ve kaynak verimliliği sağlanılabilir
The influence of trisodium citrate dihydrate complexing agent on the structural, electrical and optical properties of ?-MnS thin films
No full textgamma-MnS films were prepared using trisodium citrate (TSC) complexing agents at different molarity values (0.5, 1 and 1.5 M). Characterization techniques such as X-ray Diffraction (XRD), optical absorption spectra, Field Emission-Scanning Electron Microscope (FE-SEM), Energy Dispersive X-ray (EDX) and Hall Effect were used to determine the properties of the films. As a result of XRD analysis, it was observed that the films prepared with 0.5 M TSC were formed in amorphous structure, while the films prepared with 1 M and 1.5 M TSC were formed in polycrystalline structure and hexagonal phase. In addition, it was determined that the crystallization of the films increased with the increase in the molarity of TSC. The optical gaps of the films were determined as 3.77 eV, 3.44 eV and 3.40 eV, respectively, depending on the increase in molarity of TSC. The refractive index values of the MnS films in the visible region (400-700 nm) were calculated as 1.87, 2.02 and 2.10, respectively. Electrical resistivity and mobility values from Hall Effect measurements were measured as 1.01 x 10(6), 9.68 x 10(5), 4.10 x 10(5) ohm cm and 33.01, 36.93, 56.53 cm(2)/Vs, respectively, depending on the increase in TSC molarity value.Hakkari University [FM20LTP8]This work was supported by Hakkari University under project number FM20LTP8
VSC MT-HVDC Fault Identification Based on the VMD-TEO and Artificial Neural Network
No full text2nd International Conference on Power Systems and Electrical Technology, PSET 2023 -- 25 August 2023 through 27 August 2023 -- Milan -- 195631This paper presents a single-ended protection scheme based on an artificial neural network (ANN) and a variational mode decomposition (VMD) for the voltage source converter (VSC) multiterminal high voltage DC (MT-HVDC) system. The developed technique depends on the utilization of VMD for feature extraction of measured voltage and current signals which are transferred into the intrinsic mode function (IMF). Then the Teager energy operator (TEO) is applied for tracking the energy content of the IMFs. The energy content of the IMFs is fed into the multilayer perceptron neural network (MLP) neural network for the identification of faults in a test system. The effectiveness of the proposed method is verified on a four-terminal HVDC system with several case studies including high impedance faults (HIFs), and disturbances by using the PSCAD/EMTDC software. The developed ANN-based fault identification method provides promising results in terms of accurate and fast fault detection features. Simulation results verify the accuracy of the proposed protection method for VSC MT-HVDC systems. © 2023 IEEE
THE RELATIONSHIP BETWEEN SUSTAINABILITY AND EARNINGS MANAGEMENT: A STUDY ON OECD COUNTRIES
No full textThis study aims to investigate the relationship between earnings management and sustainability performance of publicly traded (other than financial companies) companies in OECD countries. We analysed this in two periods, specifically 2000-2009 and 2010-2020 in founding member countries of OECD by using Panel Data Analysis techniques. The increase in sustainability performance data led to a natural break point in the analysis. The findings show that there is a statistically significant relationship between sustainability performance and earnings management with all its sub-components revealing that an increase in sustainability performance leads to a decrease in earnings management
Meta-sezgisel algoritmalar kullanılarak vitamin eksikliği tahminlemesi
No full textLisansüstü Eğitim Enstitüsü, Bilgisayar Mühendisliği Ana Bilim Dalı, Bilgisayar Mühendisliği Bilim DalıKan, insanoğlunun varoluşundan beri büyük bir öneme sahip olmuştur. Antik çağlarda bile insanlar, hastalıkların ve sağlık durumlarının belirlenmesi için kanın önemli bir unsur olduğunu biliyorlardı. Tıp alanındaki gelişmeler ve teknolojik ilerlemeler sayesinde, kan testleri insanların sağlık durumlarını anlamak için önemli bir araç haline gelmiştir. Bu nedenle kan analizi, tıbbi teşhis ve tedavi süreçlerinin temel bir parçası haline gelmiştir. Bu çalışmada, B12 vitamini tahminlemesi üzerine bir proje gerçekleştirilmiştir. Veri seti, 2-92 yaş aralığında 509 kadın, 398 erkek olmak üzere toplam 907 kişinin kan değerlerini içermektedir. SVR, LASSO ve Ridge regresyonlarında, regresyon modellerinin performansını artırmak ve aşırı uyma (overfitting) problemini azaltmak amacıyla kullanılan modellere ait parametreler optimize edildi. SVR için best 'C' ve gamma değerleri sırasıyla 0.072 ve 0,107 olarak bulundu ve MAE, MSE, RMSE ve ??2 değerleri için optimizasyon sonrasında sonuçlar 0.145, 0.024, 0.152 ve 0.890 olarak iyileşti. Lasso regresyon için best 'alpha' değeri 0.011 olarak bulundu. MAE, MSE, RMSE ve ??2 değerleri için optimizasyon sonrasında sonuçlar 0.030, 0.003, 0.062 ve 0.983 olarak iyileşti. Ridge regresyon için best 'alpha' değeri 0.013 olarak bulundu. MAE, MSE, RMSE ve ??2 değerleri için optimizasyon sonrasında sonuçlar 0.063, 0.079, 0.289 ve 0.623 olarak iyileşti.Bu çalışma, tıp alanında kan analizlerinin önemini ve regresyon analizinin tahminleme gücünü vurgulamaktadır.Blood has held significant importance throughout human existence. Even in ancient times, people recognized the crucial role of blood in determining diseases and health conditions. With advancements in the field of medicine and technological progress, blood tests have become a vital tool for understanding individuals' health. Therefore, blood analysis has become an integral part of medical diagnosis and treatment processes. In this study, a project was conducted on the prediction of B12 vitamin levels. The dataset includes blood values of a total of 907 individuals, ranging from 2 to 92 years, with 509 females and 398 males. Parameters of models used to enhance the performance of regression models and reduce the risk of overfitting in SVR, LASSO, and Ridge regressions were optimized. For SVR, the best 'C' and gamma values were found to be 0.072 and 0.107, respectively, resulting in improved outcomes with MAE, MSE, RMSE, and ??2 values of 0.145, 0.024, 0.152, and 0.890 after optimization. The optimal 'alpha' value for Lasso regression was 0.011, leading to enhanced results post-optimization with MAE, MSE, RMSE, and ??2 values of 0.030, 0.003, 0.062, and 0.983. In Ridge regression, the best 'alpha' value was determined as 0.013, and the results improved after optimization with MAE, MSE, RMSE, and ??2 values of 0.063, 0.079, 0.289, and 0.623. This study emphasizes the importance of blood analysis in the medical field and the predictive power of regression analysis