itüdergisi (E-Journals - İstanbul Teknik Üniversitesi, Istanbul Teknik Uniersity., İTÜ)
itüdergisi (E-Journals - İstanbul Teknik Üniversitesi, Istanbul Teknik Uniersity., İTÜ)Not a member yet
1059 research outputs found
Sort by
Mezo-gözenekli MCM-41'in sıcak gazlardan kükürt giderme sorbentlerinde inert taşıyıcı olarak kullanılması
Get PDFGases from gasification processes mostly contain H2S, which is a very corrosive and poisonous gas, causes serious problems in applications where they are used. Thus, in many cases it is necessary to reduce the H2S concentration in these gases from percent to ppm level. For example, the allowable H2S concentration for the combine cycle power plants is 20-100 ppm while it is required to be lower than 1 ppm for the fuel cell systems. Removal of H2S form gases is a serious problem and needs to be solved for the fuel cell applications. A lot of studies have been devoted to this subject. In these studies, commonly metals and metal oxides were investigated as adsorbent material for this purpose. It was found from the studies that the suitable metals are Ca, Co, Cu, Fe, Mn, Mo and Zn. Cu and Zn, with their superior properties, are classified as most favorable metals. The pure metal oxides used as adsorbents, however, suffer from evaporation, loss in the surface area and porosity, sintering and mechanical disintegrations that affect their performance and life time adversely. For example, there is a loss problem due to evaporation for metallic Zn at elevated temperatures or there is a surface area loss problem for porous Cu. In order to overcome these problems, pure metal oxides or their admixture are loaded to inert carriers/supports with high surface areas. These are composite materials that consist of an active metal oxide(s) and an inert carrier. The main properties required for carrier materials are to be inert, high surface area, channels of large diameters, good mechanic strength, no phase change during sulfidation-regeneration process and hydrophobicity. Some inert oxides such as Al, Ti, Zr oxides and high silica zeolites are the most commonly used carriers. In this study, an adsorbent was prepared by using copper and mesoporous silica material for the H2S removal from hot gases. CuO was used as the active component and mesoporous MCM-41 was used as the carrier materials. CuO/MCM-41 type adsorbent was prepared by wet impregnation method, in which copper was loaded to the MCM-41 mesoporous material. The silica based carrier was also synthesized as a part of the study. By using wet impregnation method, one CuO/MCM-41 type adsorbent sample was prepared. The adsorbent sample contains 37.46% Cu by weight. The fresh and sulfided adsorbent samples were characterized by using X-ray diffraction (XRD), nitrogen adsorption, chemical analysis, thermal gravimetric analysis (TGA), H2-chemisorption. The H2S removal characteristics of the adsorbent sample from hot gases were investigated by using a laboratory scale fixed bed reactor of 14 mm diameter. The experimental set up consisted of three major sections: 1. gas supply, 2. reactor, and 3. gas analysis unit. Experiments were carried out as sulfidation-regeneration cycles at a temperature range of 823-873K. In these experiments, a gas mixture of H2S (2000 ppm), H2 (20%) and N2 (balance) was fed into the reactor in the sulfidation steps. Sulfidation steps were ended, when the H2S concentration of the reactor effluent reached 20 ppm which was selected as the breakthrough point. Regeneration of the sulfided adsorbent samples were carried out by gas mixtures composed of (5-10%) and N2 (balance). The results of the nitrogen adsorption analysis showed that, the adsorbent sample that was prepared by wet impregnation method, has lower BET surface areas and lower pore volumes than that of the original carrier material, MCM-41. This result indicates that the loaded CuO, partly plugs pores and pore inlets of the carrier. The decrease in the BET surface area in the pelletized samples was approximately 66% for CuO/MCM-41. The drop in pore volume was measured to be 72% for the same sample. The sample was tested for three consecutive sulfidation-regeneration cycles. The average breakthrough point sulfur capacities of the CuO/MCM-41 type adsorbent sample was determined to be 1.89 g S/100 g adsorbent, after three cycles. The XRD and nitrogen adsorption analysis indicated that the structure of the sample was distructed to some extend during the sulfudation-regeneration processes. In spite of this fact, change in the sulfur up take was not comparable and remained rather limited. Keywords: Desulfurization, MCM-41, wet impregnation.Yüksek oranda kükürt içeren fosil yakıtların gazlaştırılmasıyla elde edilen hidrojence zengin gaz karışımları birçok safsızlık içerir. Bu safsızlıkların türü ve miktarı gazlaştırılan fosil yakıtın kaynağına ve türüne göre değişiklik gösterir. Gazların içerdiği H2S, özellikle yakıt hücresi gibi uygulamalarda yakıt hücresi elektrotlarını zehirleme gibi ciddi sorunlara yol açmaktadır. Bu çalışmada, bu tür sıcak gazlardan hidrojen sülfürün (H2S) uzaklaştırılması amacıyla kullanılabilecek yenilenebilir sorbentlerin geliştirilmesi hedeflenmiştir. Yapılan çalışmalar periyodik tabloda yer alan geçiş metallerinin birçoğunun bu amaçla kullanılabileceğin göstermiştir. Ancak, bu metallerin tek başlarına kullanılmaları durumunda, kükürt tutma kapasitesinin süreç içinde önemli ölçülerde azalması, yüzey alanlarının küçülmesi, termal ve mekanik özelliklerinin yetersiz kalması gibi sorunlarla karşılaşılmaktadır. Bu sorunları çözmek için başvurulan yöntemlerden bir tanesi de bir taşıyıcının kullanılmasıdır. Bu çalışmada, mezo-gözenekli bir malzeme olan MCM-41’ın taşıyıcı olarak kullanıldığı CuO/MCM-41 tipi sorbent geliştirilmiştir. Kullanılan MCM-41 yine çalışma çerçevesinde sentezlenmiştir. Taşıyıcı malzemeye, bakır nitrat kullanılarak yaş emdirme yöntemi ile CuO yüklenmiştir. Numunenin kükürt tutma özelikleri, 2000 ppm H2S içeren, azot hidrojen karışımı kullanılarak incelenmiştir. Çalışmalar sırasında H2S için eşik nokta (breakthrough point) derişimi 20 ppmv olarak seçilmiştir. Deneysel çalışmada elde edilen sonuçlar MCM-41’in sıcak gazlardan kükürt gidermek amacıyla kullanılacak sorbentlerin hazırlanmasında taşıyıcı malzeme olarak kullanılabileceğini göstermektedir. %37.46 oranında Cu içeren sorbentin eşik nokta kükürt tutma kapasitesi 3. kükürt tutma-yenilenme döngüsü sonunda yaklaşık olarak 1.89 g S/100 g sorbent, kapasite kullanma oranı ise yaklaşık %10 olarak ölçülmüştür. Kükürt tutma-yenilenme süreci, sorbentin yapısında bazı değişmelerin meydana gelmesine neden olmakta, ancak sorbent işlevini sürdürmeye devam edebilmektedir. Anahtar Kelimeler: Sıcak gazlar, H2S giderme sorbenti, MCM-41
Epileptic seizure detection from SEEG data by using higher order statistics and spectra
Get PDFEpilepsi hastalığı kişilerin normal ve sosyal hayatını olumsuz etkileyen bazı durumlarda ölümle sonuçlanabilen ciddi bir hastalıktır. EEG (elektroensefalogram) uzmanı tarafından EEG’nin incelenerek hastanın durumu hakkında bilgi vermesi hala klinikteki altın standart olarak kabul edilmektedir. EEG verilerini bu yöntemle analiz etmek çok uzun veri kaydının incelenmesini gerektirmektedir. Ayrıca, aynı EEG verisini, farklı EEG uzmanları farklı olarak yorumlayabilmektedir. Epilepsi hastalığının EEG verilerinden tespitinde otomatik bir yöntemin gerekliliği açıktır. Bu çalışmada yüksek dereceli istatistikler kullanan ikiz-izge ve ikiz-tutarlılık yöntemleri epileptik atak içeren subdural elektroensefalogram (SEEG) verilerine uygulanarak, atak zamanları tespit edilmiştir. Epilepsi atağının doğrusal olmayan yapısı göz önünde bulundurularak bir algoritma geliştirilmiştir. SEEG verileri durağan kabul edilen zaman parçalarına bölünerek ikiz-izge yöntemi uygulanmıştır. İkiz-izgenin normalize edilmiş hali ikiz-tutarlılık hesaplanmıştır. İkiz-tutarlılık matrislerinden doğrusal olmama ölçüsünü gösterecek değişkenler belirlenmiştir. Bu değişkenler kullanılarak epileptik atak tespit edilmiştir. Epileptik atak tespitinde değişkenlerin performansı ROC (Receiver Operating Characteristic curve) kullanılarak değerlendirilmiştir. Faz kuplajı olan ikiz-frekansların sayısını gösteren değişken ile 0.1853 bağıl toplam hata oranında %53.49 duyarlılık ve %90.25 özgüllük elde edilmiştir. Diğer bir doğrusal olmayan yöntem Lempel Ziv karmaşıklık ölçüsü ve doğrusal bir yöntem olan güç izge analizi aynı işaretlere uygulanmıştır. İkiz-izge temeline dayalı sunulan bu algoritma ile ataklar tespit edilmesine rağmen, Lempel Ziv karmaşıklık ölçüsü ve güç izge analizi atakları tespit etmede başarılı olamamıştır. Anahtar Kelimeler: İkiz-izge, ikiz-tutarlılık, Lempel ve Ziv karmaşıklığı, subdural elektroensefalogram (SEEG), yüksek dereceli istatistikler. Epilepsy is a disease that has negative effects on human's normal and social lives. On some cases epilepsy can even be lethal. Expert interpretation on EEG (electroencephalogram) data is still the 'golden standard' in clinical applications where experts analyze the EEG data and generate the reports for the patient's situations. To generate the reports experts are obliged to examine and analyze very long EEG data in which the interpretation of the data may change from one expert to another. It is obvious that an automatic method is required to detect the epilepsy from EEG data. The starting time or early detection of the seizure is very important in terms of diagnosis, treating and controlling of the epilepsy. It is known that the noise, which is generated by recording devices and environment, has negative effects on analysis and interpretation of EEG while observing and analyzing the scalp EEG. For that reason, the works in literature are concentrated on subdural EEG (SEEG) in recent years. Subdural EEG is collected from patients, which do not answer to the medical treatment and candidate for brain surgery. The epileptic area in brain is decided by using MRI results, patient medical history, and clinical findings before medical operation. Patients who are decided that have epileptic area in their brain are taken to the medical operation. SEEG is recorded during the operation by using electrodes, which are placed on the celebral cortex. The SEEG data is loaded to the computers or CDs as digital data after required amplification and filtering. The aim of this work is to apply bispectrum and bicoherence methods of higher order statistics to epileptic seizure SEEG data and to detect the changes that occur during, before and just after the seizure. To achieve these objectives an algorithm is developed. In this algorithm the epileptic seizure is assumed as stationary and nonlinear. The first step is to apply bispectrum method to time framed EEG data segments, which is assumed to be stationary. Bicoherence, which is also called normalized bispectrum, is calculated from bispectrum. From the bicoherence matrix the parameters that are the measures for non-linearity, is determined. The values of bicoherence matrix are between 0 and 1 in theory. Whereas the values, which are showing uncoupled frequencies, might be close to zero but not every time in practice. A threshold level is used to determine true-coupled bifrequencies. The parameters that are the measures for non-linearity, is determined from the bicoherence matrix. Employing the non-linearity measures from each time frame, epileptic seizure is detected. The performance of the parameters on epileptic seizure detection is determined by ROC (Receiver Operating Characteristic curve). According to this method the parameter that shows the number of bifrequency, which is above a threshold level in bicoherence matrix, is determined as the most suitable parameter on epileptic seizure detection. Using this parameter epileptic seizure is detected. For this data package on a total error rate of 0.1853 sensitivity of %53.49 and specificity of %90.25 is achieved. Power spectrum is applied to the same signals to compare the performances of methods. Welch Method is used for estimation power spectrum. Normalized power spectrum and power in different frequency bands are used to detect the epileptic seizures. Power spectrum is unable to detect the epileptic seizures with acceptable total error rate for this data package. Lempel and Ziv Complexity is another nonlinear method. Lempel and Ziv Complexity and its variants are popular metrics for characterizing biological signals. Lempel and Ziv Complexity is also applied to the same signals to compare the performances of methods. Although Lempel and Ziv Complexity is applied to SEEG data to detect the seizures in literature it is unable to detect the epileptic seizures with acceptable total error rate for this data package. Therefore, parameters calculated from bicoherence provide a good characterization of epileptic data and detection of epileptic seizures. There is not any nonlinear method yet which is used for the clinical applications to detect epileptic seizure with high sensitivity and specificity. After the discovery of the possibility to control and treat the epileptic seizure by electrical stimulating the epileptic focus region of the brain, research to find methods for detecting the epileptic seizures are accelerated in a few years. Because the results of this study enlighten the literature, it will give an important contribution to a scientific hot topic. Keywords: Bispectrum, bicoherence, Lempel and Ziv complexity, subdural electroencephalogram (SEEG), higher order statistics
Incremental task selection and allocation for the real-world multiple travelling robot problem
Get PDFBu çalışmada, Çoklu-Robot Çoklu-Hedef Ataması problemi için bir gerçek zamanlı görev seçme ve atama yöntemi önerilmektedir. Çoklu-Robot Çoklu-Hedef Ataması problemi, literatürde iyi bilinen ve tüm veri kümeleri için en iyi çözümlerin polinomsal algoritmalarla bulunamadığı MTSP (Multiple Traveling Salesman Problem) probleminin her bir hedefin en az bir robot tarafından ziyaret edilmesini gerektiren değişik bir uyarlamasıdır. Bu problem için farklı eniyileme amaç fonksiyonları tanımlanabilir (Örn. yürütme zamanını iyileme, robotların toplam yollarını iyileme, vb.). Bu makalede, bu problemin gerçek dünya versiyonu ele alınıp, yürütme zamanında ortaya çıkabilecek problemler irdelenmiştir. Çoklu-robot sistemlerinin ortaklaşa çalışmalarında karşılaşılan en büyük güçlükler, görevlerin yürütme zamanının önceden kesin olarak tahmin edilememesi, yürütme süresince değişebilen maliyet değerleri ile belirsizlik ve tutarsızlıklardan kaynaklanır. Bu çalışmada, yürütme zamanı kısıtlamalarını da aşmak üzere etkin bir dinamik görev seçim ve atama yöntemi önerilmekte ve önerilen yöntemin, hem benzetim ortamlarında hem de gerçek robotlar üzerinde yapılan testlerle başarım analizi yapılmaktadır. Önerilen yöntem, yürütme zamanında oluşabilecek birçok hataya karşı dayanıklı olarak sistemin güncel durumuna göre artımlı ve dağıtılmış bir görev ataması gerçekler. Gerçek zamanlı yürütmeye uygun şekilde iletişim gereksinimleri minimumda tutulmaya çalışılmıştır. Gerçeklenen testler yöntemin bilgi-işlemsel açıdan etkin ve düşük maliyetli, sistemin hatalara karşı dayanıklı olmasını sağlayacak şekilde çalıştığını gösterir niteliktedir. Anahtar Kelimeler: Çoklu-robot sistemleri, gerçek zamanlı çoklu-robot çoklu-hedef atama, artımlı görev seçimi, dinamik görev atama, hataya dayanıklılık.In this study, the real-world Multiple Travelling Robot Problem (MTRP) is analyzed and an integrated approach is proposed to solve this problem in real time. MTRP is a generalization of the well-known Multiple Traveling Salesman Problems and is solved by a multi-robot team. In MTRP, a different version of the well known NP-hard MTSP (Multiple Traveling Salesman Problem), each target must be visited by at least one robot in its open tour. Depending on the selected application domain, various objectives may be defined for this problem (e.g. minimization of total path length, time, makespan etc.). Although this problem has emerged from Operations Research, it has become one of the main problems in multi-robot research for different missions (e.g., search and rescue, space, or surveillance operations, etc.). In this article, besides the target allocation issue, the real challenges of this problem are presented. Unpredictability of the exact processing times of tasks, unstable cost values during runtime and inconsistencies due to uncertain information form the main difficulties of the task allocation problem for robot systems. Since the real world is beyond the control of robots, in most cases, the Operations Research solutions are not directly applicable due to either robot hardware/software limitations or environmental dynamics. These approaches may become impractical when the size of the mission is even moderate or the cost values change frequently because of the uncertain knowledge, changes in the environment (including failures) or the changing structure of the mission (e.g. online tasks). Furthermore, robots have continuous path planning burdens for target sets in dynamic environments. Expensive computational efforts for initial allocations may become redundant. We propose a solution as a generalized framework " DEMiR-CF (Distributed and Efficient Multi-Robot Cooperation Framework) " which includes dynamic task selection, distributed task allocation and contingency handling mechanisms. These mechanisms and lower level robot controllers, the motor and sensory modules are integrated together to solve the real-world MTRP. DEMiR-CF, while capable of handling diverse contingencies, performs an incremental task allocation method based on the current information about the environment. Globally efficient solutions are obtained by the proposed mechanisms that form priority based rough schedules and select the most suitable tasks from these schedules. Rough schedules are formed by using information regarding the beliefs about the other robots. Since DEMiR-CF is for real-world task execution, communication requirements are kept at minimum as much as possible. The approach is distributed and computationally efficient. Target allocation and route construction is integrated into each other by an incremental assignment approach. Real time conditions and contingencies that change the problem instance are handled at the same time by the designed contingency handling mechanisms. Robots keep system models to correct models of their own or warn other robots to maintain system consistency. Empirical evaluations of the system performed on the Webots simulator and on Khepera II robots reveal the efficiency of the integrated components of the approach. Experiments are designed in three sets. In the first set of the experiments, evaluations of the proposed cost functions to be used with DEMiR-CF are performed. Comparisons are made both with the optimal results taken from the IP solver CPLEX and that of the Prim Allocation approach, one of the efficient methods to solve this problem. As the results illustrate, allocating all targets from scratch and generating routes of robots may result in suboptimal solutions. Therefore the target allocation and the route construction should be integrated for efficient heuristic approaches. This integration and incremental allocation is also useful for eliminating redundant calculations especially in highly dynamic or unknown environments. In the second set of experiments, scalability of the framework and the response efficiency of the contingency handling mechanism integrated into the framework are evaluated. The scalability of the approach is validated and the efficiency of using the contingency handling mechanisms is observed in the results. The third set of experiments is performed on real robots. As results illustrate, both incremental task selection and allocation approaches produce efficient results and the contingency handling mechanisms make the system robust and allow the handling of real-time online situations. Keywords: Multi-robot systems, multi-robot multi-target assignment in real-time, incremental task selection, dynamic task allocation, robustness.
Border feature detection and adaptation for classification of remote sensing images
Get PDFGeleneksel görüntü işleme tekniklerinin direkt olarak uzaktan algılamaya uygulanması, sadece multispektral datalar için geçerli olabilir. Öznitelik vektörü boyutu 100-200 civarında olan hiperspektral dataların analizi için gelişmiş algoritmalara ihtiyaç vardır. Bununla birlikte, uzaktan algılamada, genellikle sınırlı sayıda eğitim örneğinin olması, özellikle öznitelik vektörünün boyutunun büyük olduğu hiperspektral datalarda, parametrik sınıflayıcıların kullanımını kısıtlar. Bu çalışmanın amacı, istatistiksel dağılıma bağlı olmayan, sadece eldeki eğitim örneklerine dayanan bir algoritma geliştirerek yukarıda özetlenen uzaktan algılama için genel sınıflandırma problemlerinin üstesinden gelmektir. Önerilen Sınır Özniteliklerinin Belirlenmesi ve Adaptasyonu (SÖBA) algoritması, karar yüzeylerine yakın sınır öznitelik vektörlerini kullanır ve bu sınır öznitelik vektörleri, maksimum marjin prensibini sağlayacak şekilde adapte edilerek, öznitelik uzayında doğru bölütlemenin yapılmasını sağlar. SÖBA algoritması iki bölümden oluşur. İlk aşamada sınır öznitelik vektörlerinin başlangıç değerleri uygun eğitim kümesi elemanlarından, yönetilebilir sayıda atanır. Daha sonra uygulanan adaptasyon işlemiyle, öğrenme süreci gerçekleştirilerek sınır özniteliklerinin, sonuç değerlerine ulaşması hedeflenir. Sınıflandırma sonuç sınır öznitelik vektörlerine olan en yakın 1 komşuluk (1-EK) kuralı uyarınca yapılır. Ek olarak, SÖBA algoritmasının sınır öznitelik vektörlerinin başlangıç değerlerine ve eğitim kümesi elemanlarının eğitimde kullanılma sırasına bağlı olarak her çalışmasında kabul edilebilir derecede farklı sınır karar yüzeyleri oluşturması, konsensüs yapılarda kullanılması için elverişli bir özelliktir. Böylece birçok defa çalıştırılan SÖBA kararlarının uygun kurallarla birleştirilmesiyle tek bir sınıflayıcının aldığı karardan çok daha doğru kararlar elde edilebilir. Anahtar Kelimeler: Uzaktan algılama, hiperspektral data sınıflandırma, konsensüs.Various types of sensors collect very large amounts of data from the earth surface. The characteristics of the data are related to sensor type with its own imaging geometry. Consequently, sensor types affect processing techniques used in remote sensing. In general, image processing techniques used in remote sensing are usually valid for multispectral data which is relatively in a low dimensional feature space. Therefore, advanced algorithms are needed for hyperspectral data which have at least 100-200 features (attributes/bands). Additionally, the training process is very important and affects the generalization capability of a classifier in supervised learning. Enough number of training samples is required to make proper classification. In remote sensing, collecting training samples is difficult and costly. Consequently, a limited number of training samples is often available in practice. Conventional statistical classifiers assume that the data have a specific distribution. For real world data, these kinds of assumptions may not be valid. Additionally, proper parameter estimation is difficult, especially for hyperspectral data. Normally, when the number of bands used in the classification process increases, precise detailed class determination is expected. For high dimensional feature space, when a new feature is added to the data, classification error decreases, but at the same time, the bias of the classification error increases. If the increment of the bias of the classification error is more than the reduction in classification error, then the use of the additional feature decreases the performance of the decision algorithm. This phenomenon is called the Hughes effect, and it may be much more harmful with hyperspectral data than with multispectral data. Our motivation in this study is to overcome some of these general classification problems by developing a classification algorithm which is directly based on the available training data rather than on the underlying statistical data distribution. Our proposed algorithm, Border Feature Detection and Adaptation (BFDA), uses border feature vectors near the decision boundaries which are adapted to make a precise partitioning in the feature space by using maximum margin principle. The BFDA algorithm well suited for classification of remote sensing images is developed with a new approach to choosing and adapting border feature vectors with the training data. This approach is especially effective when the information source has a limited amount of data samples, and the distribution of the data is not necessarily Gaussian. Training samples closer to class borders are more prone to generate misclassification, and therefore are significant feature vectors to be used to reduce classification errors. The proposed classification algorithm searches for such error-causing training samples in a special way, and adapts them to generate border feature vectors to be used as labeled feature vectors for classification. The BFDA algorithm can be considered in two parts. The first part of the algorithm consists of defining initial border feature vectors using class centers and misclassified training vectors. With this approach, a manageable number of border feature vectors is achieved. The second part of the algorithm is adaptation of border feature vectors by using a technique which has some similarity with the learning vector quantization (LVQ) algorithm. In this adaptation process, the border feature vectors are adaptively modified to support proper distances between them and the class centers, and to increase the margins between neighboring border features with different class labels. The class centers are also adapted during this process. Subsequent classification is based on labeled border feature vectors and class centers. With this approach, a proper number of feature vectors for each class is generated by the algorithm. In supervised learning, the training process should be unbiased to reach more accurate results in testing. In the BFDA, accuracy is related to the initialization of the border feature vectors and the input ordering of the training samples. These dependencies make the classifier a biased decision maker. Consensus strategy can be applied with cross validation to reduce these dependencies. In this study, major performance analysis and comparisons were made by using the AVIRIS data. Using the BFDA, we obtained satisfactory results with both multispectral and hyperspectal data sets. The BFDA is also a robust algorithm with the Hughes effect. Additionally, rare class members are more accurately classified by the BFDA as compared to conventional statistical methods. Keywords: Remote sensing, hyperspectral data classification, consensual classification
The behavior of a thin steel plate shear wall with semi-rigid beam-to-column connections under cyclic loads
Get PDFİnce levhalı çelik perdeler, büyük başlangıç rijitliğine ve yüksek süneklik ile birlikte büyük enerji sönümleme kapasitesine sahip olmaları nedeniyle oldukça etkili yatay yük taşıyıcı sistemlerdir. Çok katlı, levhalı bir çelik perde, ince çelik levhalar ile bu levhaların bağlandığı kolon ve kiriş elemanlardan oluşmaktadır. Bu araştırma kapsamında, yarı-rijit kiriş-kolon birleşimleri içeren ince levhalı çelik perdelerin tersinir yükler altında davranışlarını anlayabilmek amacıyla deneysel bir çalışma yürütülmüştür. Deney numunesi olarak 1/3 ölçekli, tek katlı ve tek açıklıklı çelik perde kullanılmaktadır. Kolon eksenleri arasındaki mesafe 1800 mm, kiriş eksenleri arasındaki yükseklik ise 1200mm'dir. İnce levha kalınlığı 0.50 mm olarak belirlenmiştir. Deney, tersinir yükleme altında ATC-24 esas alınarak gerçekleştirilmiştir. Numunenin deney sırasında sergilediği davranışı tanımlayabilmek amacıyla 53 adet şekil değiştirme ölçer ile 12 adet yer değiştirme ölçerden yararlanılmaktadır. Deney sonucunda sisteme ait, başlangıç rijitliği, akma dayanımı ve akma yer değiştirmesi gibi karakteristik bilgiler ile yük-yer değiştirme eğrisi elde edilerek ince levhalı çelik perdenin çevrimsel davranışı değerlendirilmiştir. Ayrıca sistemin davranışını anlayabilmek amacıyla, deney sırasında belirli yer değiştirme seviyelerinde ayrıntılı olarak yapılan tespit ve gözlemler de sunulmuştur. Ayrıntılı tespit ve gözlemler ile deney sonuçlarının değerlendirmesi sonunda, göçmenin ince levhanın üst kenarının birleşim kesiti boyunca yırtılması sonucu meydana gelmesine rağmen, levhanın sistemin enerji sönümleme kapasitesine büyük oranda katkıda bulunduğu görülmüştür. Böylece ince levhalı çelik perdenin alternatif bir yatay yük taşıyıcı sistem olarak kullanılabileceği düşünülmektedir. Anahtar Kelimeler: İnce levhalı çelik perde, tersinir yükleme, yarı-rijit kiriş-kolon birleşimi.Thin steel plate shear walls are very effective and economical lateral load-resisting systems. A thin steel plate shear wall is composed of vertical steel plates which are named as infill panels and columns and beams connected to steel panels. The beam-to-column connections may be either full moment-resisting and partially moment-resisting (semi-rigid) joint types or simple joint type. Infill steel panels are generally attached to the surrounding frames by either fillet welds or bolts. Utilization of steel plate shear walls began in early 1970's. Initially, stiffened steel shear walls were used in Japan in new construction and in the USA for seismic retrofit of the existing buildings as well as in new buildings. Early steel shear walls have been designed such that shear buckling of the infill plate should not be occurred. Thus, in order to prevent the shear buckling of the steel panels vertical and horizontal stiffener members attached to the surfaces of the steel infills were used. In that design philosophy, post-buckling behavior of the steel panel was neglected. In this study, an experimental research which includes one, 1/3 scale, one-storey, single bay test specimen, so-called SW-A-H is performed under cyclic loading in order to understand the cyclic response of thin steel plate shear wall systems. The test specimen used for this purpose consists of a boundary frame and thin steel plate which is connected to boundary frame members on all edges. Infill plate is 0.50mm thick and has a height of 900mm and width of 1520mm. Self drilling screws with 5.5mm diameter are utilized to provide the connection between the infill and surrounding frame members. Also, a fish plate welded to the flange of each frame member is used. In this experimental study, the specimen is tested in accordance with the ATC-24, Guidelines for Cyclic Seismic Testing of Components of Steel Structures, by Applied Technology Council. In accordance with ATC-24, the number of cycles with a peak deformation less than yield displacement were eighteen. The number of cycles with peak deformation that is equal to yield displacement are three. After the cycles with peak deformation larger than yield displacement, the increment of yield displacement was applied to the peak displacement for each step. In order to determine the material properties of the infill panel and beam end plates tension coupon tests were conducted at Construction Materials Laboratory of ITU. The coupons were cut in two orthogonal directions from the infill plate. While six coupons were used for the infill panels three coupons were taken for the beam end plates. Cyclic test of the specimen was performed at Structural and Earthquake Engineering Laboratory of ITU. A loading system with an actuator which is capable of producing 250kN and having a stroke with the capacity of ±300mm is utilized. After the experimental observations during testing of the specimen are described, the data from the measurement devices such as strains, displacements and loads are evaluated to generate the information about the behavior of the specimen and response of the infill plate. Therefore, several properties of systems are extracted from the experimental data, namely, initial stiffness, yield base shear and displacement, cumulative energy dissipation and displacement ductility. Furthermore, the orientation and the magnitude of the principal stresses and strains in the infill plate during linear and nonlinear portions are examined and the variation of the strain magnitude and energy dissipation across the infill is investigated. The results obtained from the tests of the specimen, load-deflection curve and envelope curves of the hysteresis, are also presented. The use of thin steel plate shear wall to resist seismic forces is possible. The experimental results are shown that the entire infill panel of the shear wall specimen SW-A-H participated in dissipating energy. The specimen utilizing flat infill panels and screwed connections to the surrounding frame is found to be reasonably ductile and failure is the result of fractures (tears) in the net sections. Despite these fractures in the net sections, which appeared first at -5dy in the specimen SW-A-H the specimen did not suffer a significant loss of strength until 6 times the yield displacement. But, at the displacement level of 7dy, the shear force of SW-A-H was suddenly dropped by 34%. Keywords: Thin steel plate shear wall, cyclic loading, semi-rigid beam-to-column joint
Predicting unemployment rates with the use of cognitive mappingmethodology and Artificial Neural Networks
Get PDFÖngörü modellemesi, makro politikaların oluşturulmasında önemli bir karar destek aracı olarak görülmektedir. Bilgisayar teknolojisindeki ilerlemeler sayesinde yapay zeka sistemlerinin karar destek araçları olarak kullanımları da gün geçtikçe artmaktadır. Yapay zeka tekniklerinden Yapay Sinir Ağları yöntemi, öngörü çalışmalarında kullanılabilecek ümit veren bir yöntem olarak araştırmacılar tarafından ilgi görmektedir. Bu çalışmada Bilişsel Haritalar yöntemi ile Yapay Sinir Ağları yöntemleri kullanılarak bir öngörü modeli kurulması çalışması gerçekleştirilmiştir. Çalışmada öngörü veri seti olarak Türkiye’de işsizlik oranları verisi kullanılmıştır. İşsizlik problemi, dünya devletleri için olduğu kadar Türkiye için de en önemli sorunlar arasında yer almaktadır. İşsizliği etkileyen faktörlerin belirlenmesi ve bu faktörler kullanılarak işsizlik tahminleri yapılması, işsizlik sorununa çözüm için hazırlanacak politikaların geliştirilmesinde karar destek bilgisi oluşturulmasına yardımcı olacaktır. Çalışmanın birinci aşamasında bilişsel haritalar yöntemi kullanılarak daha önce Türkiye ile ilgili makroekonomik çalışmaları bulunan beş akademisyenin görüşü alınıp Türkiye’de işsizliği etkileyen faktörler belirlenmiştir. Uzmanlar tarafından belirlenen on bir faktör, çalışmanın ikinci aşamasında çok katmanlı Yapay Sinir Ağı kullanılarak oluşturulan çok değişkenli bir öngörü modeline girdi olarak kullanılmıştır. Çalışmada 1988 ve 2004 yılları arasında dört dönemlik veriler kullanılmıştır. En iyi öngörü modeli, oluşturulan 24 değişik model arasından seçilmiştir. Çalışma sonunda en iyi tahmin modellerine mevsimsellikten arındırılmış veri seti ile ulaşıldığı görülmüştür. Çalışmada aynı zamanda çıktı katmanında kullanılan doğrusal ve doğrusal olmayan aktivasyon fonksiyonlarının, ağ performansına belirgin bir etkisi olmadığı, mevsimsel arındırma ve (0.1;0.9) ölçek aralığı kullanılmasının ise ağ öngörü performansını olumlu etkilediği görülmüştür. Anahtar Kelimeler: Bilişsel haritalar, Yapay Sinir Ağları, işsizlik.Forecasting is accepted as an important tool in the development and application of macro policies. Considering the fact that accurate forecasts help decision makers take better decisions, forecasting studies have started to take growing interest of researchers. By the developments in the computer specifications, artificial intelligence techniques have gained more attention as decision support tools. Among the artificial intelligence techniques, artificial neural networks (ANNs) are seen as promising techniques especially in forecasting applications. In this study, cognitive mapping methodology which is used in capturing the cause-effect relationships in complex causal systems and facilitate understand the interconnections within the elements of the systems by gathering expert knowledge is used in the identification of the factors that affect unemployment in Turkey and ANNs methodology are integrated in order define a prediction model framework. The framework is applied to unemployment rate data sets in Turkey in order to understand the important factors that affect unemployment and to forecast unemployment rate in Turkey. Understanding the factors that affect unemployment and developing forecasting models by the use of these factors can help the decision makers as a decision support mechanism in defining policies to overcome unemployment problem. In the first phase of the study Cognitive Mapping methodology is used in the identification of the factors that affect unemployment in Turkey by taking the views of five different experts who have made research on macroeconomic problems of Turkey. The eleven variables identified by the experts are used in the second phase of the study as the input data set in developing a multi-variate forecasting model by using ANNs. The research period covers 1988 and 2004. Quarterly data is used in the analysis. To find the best network to forecast the output variables, a design of experiment is made and two different output activation functions; (linear function and nonlinear tangent sigmoid function), two different scaling ranges (-0.9; 0.9 and 0.1; 0.9), seasonal and deseasonalized data is used in the study. Additionally, three different input data structures are used in the study. In the first data structure, all variables identified by the experts are used in the analysis. In the second structure, unemployment rate data in one past period is added to the 11 input variables. In the third input structure, all input variables and unemployment rate are used in their lagged values as the inputs to the network. Therefore a total of 24 different models are tested. The networks are trained using MATLAB software, TRAINGDX training algorithm. To train each network, different numbers of hidden units are used and almost 85 % of the data set is used as the training set. Each run is replicated ten times with different initial weights in order to avoid getting stuck in local minima. Since the data set was not large enough to use a validation set, training is stopped at epoch 1000. The results of the study show that the best network found uses the level values in deseasonalized data sets; a tangent sigmoid function as the output activation function and a scaling range of (0.1;0.9). The structure of the input set is the second structure which includes all the 11 variables identified by the expert and the one past period value of the dependent variable itself. The results show that the specified best neural network model is able to perform satisfactorily within the test set. The test MSE is found to be 0.0015 where the training MSE is 0.0029. Furthermore in the study, the effects of the different activation functions and different scaling ranges and deseasonalization on network prediction performance are tested. The results showed that models using a scaling range of (0.1;0.9) give significantly better average MSE results compared to the models using a scaling range of (-0.9;0.9). For the output activation functions, the test results reveal the mean performances of the models using tangent sigmoid function are not different from each other statistically. The test results also indicate that deseasonalization significantly improves network performance in unemployment rate data. Finally during the analysis it is observed that the networks tend to get over trained as the number of the hidden units increase; this result is in accordance with the findings in previous research on neural networks. The results of the study indicate that ANNs may be considered as promising tools in macroeconomic forecasting studies. Keywords: Cognitive maps, artificial neural networks, unemployment
Paralel kesintisiz güç kaynaklarından oluşan dağınık güç sistemlerinde aktif yük paylaşımı
Get PDFContinuity of power flow is very important for telecommunication systems, hospitals, computers and critical industrial facilities. Uninterrupted Power Supplies (UPS) have been in use for several years for this purpose. Increased power requirement of these systems has led to operation of several UPS units in parallel. However, this has brought in issues related to the control of these systems. Initially, parallel systems composed of equally rated units were built. The controllers of these systems compared the current drawn by each unit to the average current levels to minimize the error. Equal rated unit selection requirement poses an important limitation for these systems. Later on, a communication line that carries the current and frequency information was used for control purposes. The most important disadvantage of these systems is that the communication lines are open to noise and any noise related problem may affect the whole system. Also, the system becomes more complicated due to the use of communication line. This study investigates the use of power-frequency droop method, which has been used in power systems for a long time, in parallel connected, distributed power systems. Each USP unit is supposed to have its own controller. These controllers operate only on its own output current and voltage data. As there is no communication line in this system, the operation is simpler and thus, more reliable. Controllers compute the active power by using the current and voltage values, and then adjust the frequency based on the power-frequency droop characteristic. Each unit has its own power-frequency characteristic. That means the system may consist of units with different rated power levels. When the load of the system changes each unit responds immediately and contributes in proportion with their rated power levels. This enables each unit to operate at their optimum operating level. It is obvious that this method causes the frequency to deviate from the rated value. The frequency of the system can be restored to the original value by shifting the power-frequency characteristic vertically. In this work, a distributed power system consisting of two parallel UPS units is investigated. The system can be loaded at two different points. Focus of the operation is on the inverter parts. Each inverter uses Sinusoidal PWM (SPWM) method, and has an LC filter at their outputs. Inverter outputs are connected together by a tie-line inductor. This line is required to transfer power from one unit to a load located near the other unit. PWM signals of each inverter are generated by using the frequency that is determined after using the power-frequency droop method. Since each unit has a different power-frequency droop characteristic and different power rating, they also have a different PWM frequency. This leads to a phase difference between the output voltages, and to a power flow through the tie-line. After a transient period following a load change, each unit shares the load power in proportion with their rated power values. First of all a system simulation was done by using Matlab-SIMULINK. Then, the simulation results were confirmed by an experimental set-up including a DS-1103 DSP controller and two ASIPM modules. One of the inverters was chosen to be 50% higher rated to experience transient frequency deviations when the load changes. As expected, different frequency values were observed following load changes. This led to a phase difference between the units, and thus power flow through tie-line. The frequencies are equalized at the steady state, and the phase difference is fixed. Frequency restoration algorithm then takes over, and brings the frequencies back to the rated level. The two load units used in the system are equal, but the inverters supply these loads in proportion with their power ratings, meaning that one supplies 50% more power than the other one. Simulation and experimental results are in agreement. Power-frequency droop algorithm yields the synchronous operation of two units in parallel, at a higher power level. As the individual units operate near their rated power levels the efficiency is increased. Due to the lack of communication line, the control is simpler and the operation is more reliable. Parallel operation also adds redundancy to the system. Keywords: Parallel operated UPS systems, distributed power systems, load sharing, mircogrid.Bu çalışmada, iki adet üç fazlı kesintisiz güç kaynağından (KGK) oluşan bir dağınık güç sisteminde aktif güç paylaşımı incelenmektedir. Çalışmada KGK sistemlerinin, çıkış katındaki eviricileri üzerinde durulduğundan, sistemin iki adet paralel eviriciden oluştuğu varsayılmıştır. Bu çalışmada enterkonnekte güç sistemlerinde yıllardır kullanılmakta olan yük-frekans düşüm yönteminin paralel çalışan KGK birimlerine uyarlanabildiği gösterilmiştir. Bu yöntemde her bir KGK birimi kendi denetleyicisine sahiptir ve sadece denetlediği birimin çıkış akım ve gerilim bilgilerini kullanmaktadır. Bu sayede herhangi bir haberleşme hattının kullanılmasına gerek kalmamaktadır. Denetleyiciler, birimlerden aldığı akım ve gerilim bilgilerinden aktif gücü hesaplayarak, yük-frekans düşüm karakteristiğine göre belirlenen yeni frekansı, ayar değeri olarak kullanır. Her bir birimin kendine ait bir yük-frekans düşüm karakteristiği olduğundan, sistemde farklı nominal güç değerlerine sahip KGK birimleri kullanılabilir. Sistemdeki toplam yük değişimine her bir birim anında tepki vererek, değişim miktarını kendi güçleri oranında karşılar. Böylece farklı güçlerdeki birimlerin her biri optimal bir güç seviyesinde çalışır. Her bir eviriciye ait olan kontrol algoritması, düşüm karakteristiğini kullanarak, eviricilerin güçleri oranına ve sistemde bulunan yüklerin değerine göre eviricilerin frekanslarını, tespit eder. Bu frekanslar, çıkış gerilimleri arasında faz farkına yola açar ve eviricilerin sisteme verecekleri güç ve ara bağlantı hattından akacak güç akışı sağlanmış olur. Sonuçta paralel çalışma ve sistemin toplam yükünün eviricilerin güçleri oranında paylaşılması gerçekleşmiş olur. Sistemin benzetimi Matlab-Simulink paket programında yapılmıştır. Deneysel çalışmada ise ASIPM evirici modülleri ve denetleyici kart olarak da DS-1103 kullanılmıştır. Anahtar Kelimeler: Paralel çalışan KGK sistemleri, dağınık güç sistemleri, yük paylaşımı, mikro şebeke. 
Kutu kirişli köprülerin stokastik dinamik analizi
Get PDFIn this study, under the effect of strong ground motions for bridges which have multi spans, and continuous box-girders with elastomer bearings, the finite element method and dynamic analysis steps are combined and dynamic behavior of viaduct is determined depending on various soil types and damping ratios by stochastic analysis method based on random vibration. Furthermore, based on recorded ground motion, the calculations performed by stochastic methods, are obtained in order to simulate earthquake acceleration for each soil types and at bedrock level based on white noise model depending on seismic activity of zone. By the stochastic dynamic analyses of the viaduct system, the structural responses such as displacements, bending moments and shear forces are calculated. The results in frequency domain are compared with those of made by deterministic methods in time domain. Stochastic dynamic procedure is applied to the analysis of viaduct system, thus stiffness and mass matrices are obtained to establish the finite element model and to carry out dynamic analysis. The computer code BRIFEAP is developed to discreatize the viaduct system, for dynamic analyses BRIDAP, for simulation of ground motion SIMEARTHP and BRISTOCAP is developed to perform the stochastic analysis of viaduct system. In stochastic analyses, filtered white noise model is taken as basis to consider the variation of ground motions. Simulated ground motions are generated at the bedrock level for three soil types and for twenty relevant components of recorded strong ground motions for each. Programs are developed for a small viaduct model and dynamic analyses are carried out by constituting substructure procedures at the beginning. Having carried out the analyses by using developed computer programs for the system, the displacements of the nodal points and the dynamic characteristics are obtained and compared with those of the computer package SAP2000 to verify the program developed. Later, the analyses of an actual box-girder viaduct having elastomeric bearings are carried out by making additions and improvements to the programs to represent the structural system as accurately as possible. The stiffness of elastomeric bearings decreases inertia forces due to ground motion by providing flexibility to the superstructure of the viaduct. The analysis includes the governing equations and specifications of the elastomeric bearing stiffness depending on its mechanical properties. An equivalent (or reduced) model of viaduct is developed by considering the elements that only connect the selected reference nodal points to reduce the computation volume and time. A second reduction is done in the mass and stiffness matrices for decreasing the number of the freedom which governs the free vibration of the system. By performing the dynamic analysis with reduced dynamic stiffness matrices is showed that the reduced model of the system can represent the actual system with an acceptable accuracy. Since ground motions have several types of uncertainties due to their random properties, a deterministic analysis of the system is not realistic. In the present thesis, random variations are considered for the ground motions. Stochastic analysis is carried out by using the power density function of the ground motion recorded at local site. They are filtered by adopting the transfer functions of Tajimi and Penzien. For this purpose, the corresponding power spectral density functions and the constant power spectral magnitudes are obtained for various types of soils. Having performed stochastic and deterministic dynamic analyses for the viaduct without making any simplification, a simple model representing one of the middle piers is developed and the corresponding stochastic analyses are carried out. In these analyses, the single pier with elastomeric bearings at the supports is considered. The pier is assumed to be subjected to horizontal and vertical ground motions simultaneously and the recorded and simulated ground motions are taken into consideration. The variances and average values of the base shear forces and the overturning moments are obtained for various ratios of damping and for soil types. The results are compared with those of the deterministic analysis as well. By carrying out the analyses, the stochastic methods have given results compatible with those of deterministic dynamic analysis and also have been observed as its economical and procedure superiority of the prediction of randomness. Keywords: Stochastic, white noise, simulated ground motions, bridge, dynamic, finite element.Bu çalışmada çok açıklıklı değişken kutu kesitli kirişleri olan elastomer mesnetli köprüler için sonlu eleman yöntemi ile dinamik analiz adımları birleştirilerek kuvvetli yer hareketleri etkisindeki viyadüğün dinamik davranışı, farklı zemin tipleri ve sönüm oranlarına bağlı olarak rasgele titreşim kuramına dayalı stokastik analiz yöntemi ile belirlenmektedir. Ayrıca, kaydedilmiş bir yer hareketi esas alınarak stokastik yöntemlere dayalı hesaplamalar ile her zemin grubu için ve beyaz gürültü modeli esas alınarak anakaya seviyesinde benzeştirilmiş deprem ivmeleri bölgenin sismik aktivitesine bağlı olarak elde edilmektedir. Stokastik dinamik analizler ile viyadük sisteminde yerdeğiştirme, eğilme momenti ve kesme kuvvetleri gibi yapısal tepkiler için frekans ortamında hesaplanan sonuçlar, deterministik dinamik yöntemlerle zaman ortamında yapılan analizlerden elde edilen değerlerle karşılaştırılmıştır. Stokastik dinamik çözümleme yönteminin uygulanabilmesi için köprü sisteminde sonlu eleman modelinin kurulması ve dinamik analizleri gerçekleştirmek üzere bilgisayar programları geliştirilmiştir. Viyadük sisteminin ayrıklaştırılarak sonlu elemanlarla modellenmesi için BRIFEAP, dinamik analizleri için BRIDAP, kaydedilmiş bir yer hareketinin beyaz gürültü ile anakaya seviyesinde ve farklı zemin tiplerine göre benzeştirilmesi için SIMEARTHP ve viyadük sisteminin stokastik analizleri için BRISTOCAP adlı bilgisayar programları geliştirilmiştir. Stokastik analizlerde yer hareketinin değişimi, filtrelenmiş beyaz gürültü modeli ile göz önüne alınmaktadır. Kaydedilmiş kuvvetli yer hareketinin ilgili bileşeni için anakaya seviyesinde ve 3 zemin türünde 20 şer adet benzeştirilmiş yer hareketleri türetilmektedir. Yapılan çözümlemelerden stokastik yöntemlerin deterministik dinamik yöntemlerle uyumlu sonuçlar vermesinin yanısıra stokastik yöntemin rasgeleliğin tahmininde üstünlüğü ve ekonomikliği gözlenmiştir. Anahtar Kelimeler: Stokastik, beyaz gürültü, benzeştirilmiş yer hareketleri, köprü, dinamik, sonlu eleman
The study on the possible use of microgranite in glazed porcelain tile bodies as an alternative fluxing agent
Get PDFSeramik karolar, içinde yaşadığımız konutlarda yer ve duvar yüzeylerinde kaplama malzemesi olarak kullanılan inorganik malzeme içerikli ürünlerdir. Seramik karolar gelişen teknoloji ile sırlı-sırsız, duvar-yer-granit (porselen), iç-dış mekan olmak üzere çeşitli sınıflarda üretilmektedir. Bu çalışmada Bilecik-Karaköy civarındaki mikrogranitin atritör değirmende farklı sürelerde öğütülmüş halinin sırlı porselen bünyelere katılarak kullanım olanakları araştırılmıştır. Bilindiği üzere mikrogranitler ana olarak alkali feldspat, plajioklas ve kuvars minerallerinden oluşmaktadır. Holokristalen dokuda, orta kaba taneli, açık bej, gri ve beyaz renk tonlarında bulunabilen derinlik kayalarıdırlar. İlk olarak mikrogranit hammaddesi laboratuar tipi jet değirmende 20 dak. öğütülmüş, daha sonra bu çamur laboratuar tipi atritör değirmende 270 dak. öğütülerek etüvde kurutularak toz haline getirilmiştir. 270 dak. öğütülerek hazırlanmış malzeme standart sırlı granit porselen bünye içerisine %8, %12, %16 ve %20 katılarak bünyelerin endüstriyel rulolu fırınlarda 1200ºC-49 dak. pişirimleri yapılmıştır. Pişirim sonrası bünyelerin, bulk yoğunluk, pişme küçülmesi (%), su emme (%), pişmiş mukavemet (N/mm2) ve Renk L, a ve b değerleri belirlenmiş standart bünye ile mukayeseleri yapılmıştır. Bünyelerin termal davranışları DTA-TG (Diferansiyel Termal Analiz-Termogravimetri) analizi, Faz analizleri ise XRD (X-ray Difraktometre) ile belirlenmiştir. Tüm bünyelerin mikroyapısal incelemeleri SEM (Taramalı Elektron Mikroskobu) ile yapılmıştır. Sonuç olarak çok ince öğütülerek bünyelerde kullanılan mikrogranit malzemesinin çok iyi bir ergitici malzeme olduğu ve bünye içerisinde kullanılan diğer ergitici malzemelerin oranını %60’lara yakın azalttığı ortaya çıkmıştır. Anahtar Kelimeler: Mikrogranit, atritör değirmen, feldspat, pişirim, elektron mikroskobu.The ceramic tile industry is being progressively moving its worldwide production toward new material with improved aesthetic and technical properties, e.g. glazed porcelain tiles that are able to successfully compete with ornamental stones and other building products. Porcelain tiles thanks to their superior properties such as whiteness, mechanical and frost resistance, low water absorption etc. attracted the attention of ceramic tile producers, causing the researches in this area to start in 80’s. Today the porcelain tiles are divided into three main groups; natural, polished and glazed porcelain tiles. Porcelain tiles usually posses water absorption of less than 0.5% but in many cases it is less than 0.01%, excellent wear resistance and high breaking strength which make them high-tech products. Moreover easily cleanable and anti-slip surfaces widen their applications both in business and home residences. They can be produced in either way; glazed or unglazed, coloured with traditional stains and fired at 1200-1220 ºC for 40-80 min. Porcelain tile is a highly vitrified ceramic material produced from a body formulated by mixtures of clay, quartz and sodium-potassium feldspar in which vitrification indicates a high degree of melting on firing which confers low porosity and high glass content. Feldspar minerals are essential components in igneous, metamorphic and sedimentary rocks, amongst the numerous rocks in which they are present. Feldspar minerals are essential components in igneous rocks as granite, which contains up to 50% or 70% of alkaline feldspar. Granite, however is rarely used for its feldspathic content. Rather a whole range of rocks geologically connected to granite is preferred. In this study, microgranite samples were collected from a large ceramic production region in Karaköy located in the Southeast part of Bilecik province in Turkey (NW Turkey). Due to its high alkaline (Na2O + K2O ≈ 8.0-8.5) and low Fe2O3, TiO2 content, it is likely to consider microgranite as a fluxing agent for porcelain tile bodies. The XRD pattern taken from the as-received microgranite raw material revealed the fact that the samples consisted of quartz (SiO2), albite, sanidine, muscovite and kaolinite minerals, which was supported from mineralogical point of view by DTA-TG analysis’s results as well. In the first part of the experimental study, both as-received and 270 min. attrition milled microgranite samples were added into the glazed porcelain body in the following proportions 8%, 12%, 16%, 20% meanwhile albite was removed from the standard body formulation. Following the wet ball milling of standard body and studied porcelain bodies, some test samples were prepared according to standard test procedure. The prepared samples were then taken to fast firing process in roller kilns under industrial conditions of 1200 ºC and 49 min. Prior to firing process slip properties of density (g/l), flow rate (s) and sieve residue (+63µ) were measured. Following this, the properties of fired body samples such as bulk density, shrinkage (%), water absorption (%), dry and fired strength and colour measurements expressed as L, a and b were measured as well. When compared with standard body, the results showed that the increase in microgranite amount decreased the water absorption and increased shrinkage and fired strength. Non-contact dilatometric analysis, Differential thermal analysis- thermogravimetric analysis (DTA-TGA) and X-ray diffraction (XRD) analysis were performed to determine the thermal expansion behaviours of the bodies, thermal behaviours of the bodies and the phases, respectively. Scanning electron microscope (SEM) in combination with energy dispersive X-ray spectroscopy (EDS) was further employed to observe the microstructure of the fired microgranite samples with respect to firing temperature. Moreover after firing process under industrial conditions quartz, plagioclase and mullite phases were detected in the bodies as the main phases. These observations were supported by DTA-TGA analyses as well. The inspection result of SEM image of the etched microgranite samples revealed that the sharp corners of the quartz grains were rounded by surface tension forces and the start of their partial solution into the glassy matrix has been observed. Based on these analyses and observations made during this study, it has been concluded that very fine grained microgranite material can be used in glazed porcelain tile bodies as a very good fluxing agent which leads to the suggestion that the removal of albite taken from Çine-Aydın region from the body recipe can be taken into consideration. Keywords: Microgranite, attrition mill, feldspar, firing, electron microscope
Investigation of the change in horizontal velocity gradient on the vertical axis caused by wave motion in vegetated areas
Get PDFEkolojik açıdan büyük önem arz eden sazlıklı bölgeler (sulak alanlar) hassas bir dengeye sahiptir. Özellikle göç eden kuşlar için önemli bir uğrak ve dinlenme yeri olan bu bölgeler, aynı zamanda içerdikleri zengin besin olanaklarıyla kara ve su canlıları için de önemli bir yaşam alanıdır. Bu bölgelerin korunabilmesi ve sürdürülebilir kullanımı için, mevcut hidrodinamik yapının bilinmesi gereklidir. Sazlıklı bölgelerde gerçekleştirilecek sıcak su deşarjlarıyla ortama eklenen atık suyun davranışının belirlenerek, olumsuz etkilerinin en aza indirilebilmesi bir gerekliliktir. Bu çalışmanın amacı sazlıklı ortama verilen sıcak su deşarjının dalga etkisiyle değişen hız dağılımının tespit edilmeye çalışılmasıdır. Bu amaçla, dalga yüksekliği, başlangıç sıcaklık farkı, deşarj edilen debi ve sazlık olup olmaması bağımsız değişken olarak kabul edilmiş ve bu şartlarda 17 adet kontrollü deney gerçekleştirilmiştir. Ölçülen parametreler; sıcaklık, hız, dalga yüksekliği ve dalga periyodudur. Deneysel çalışma 22×1×0.5 m boyutlarında bir kanalda gerçekleştirilmiştir. Akım hızları Vectrino+ akustik doppler hızölçer ile 50 Hz sıklığında kaydedilmiş ve en düşük kayıt süresi 1 dakika olarak belirlenmiştir. Sıcaklıklar Pt100 tipi 19 adet termometre ile ölçülmüştür. Sonuç olarak hem sazlıklı hem de sazlıksız ortamlar için dalga yüksekliği-hız dağılımı bağıntıları istatistiksel bir yaklaşımla elde edilmiştir. İstatistiksel yaklaşımda çoklu regresyon analiz yöntemi kullanılmıştır. Söz konusu bağıntılar bağıl sıcaklık, boyutsuz hız ve bağıl derinlik parametrelerinin fonksiyonları olarak verilmiştir. Elde edilen fonksiyonların istatistiksel açıdan yeterli hassasiyette olduğu görülmüştür. Sazlıksız ortamdaki hız değerleri klasik yöntemlerle hesaplandıktan sonra elde edilen bu fonksiyonlarla karşılaştırılarak sazlıklı ortamdaki dağılım elde edilebilmektedir. Anahtar Kelimeler: Sıcak su deşarjı, sazlıklı ortam, hız dağılımı. Ecologically important vegetated areas bear a tender equilibrium. These areas are specifically places for birds and the great habitats for animals as well, since these places are very wealthy when it comes to nutrients. Hydrodynamic structures of these places should be known in order to maintain a sustainable development. Thus, this study is done to reveal the hydrodynamic structures of vegetated areas and the behaviour of waste water with thermal water discharges, and eventually to reduce the negative effects of this phenomenon. The changes in the characteristics of thermal water discharges which are induced by vegetation at the receiving environment are examined with the help of an experimental study. Velocity and dispersion coefficients were tried to be determined with previously conducted studies. In addition, drift forces caused by the changes occurred in the flow by one plant or a plant group were tried to be determined. In this study, wave height, temperature difference, flow rate of the discharge, and the existence of vegetation are chosen as independent variables, and 17 sets of experiments are conducted. Temperature, velocity, wave height and wave period are the measured parameters during the experiments. Experiments are conducted in a 22×1×0.5 m dimensioned flume. Flow velocities are measured at least record is 1 minute long. Furthermore, temperatures are recorded with 19 "Pt100" type thermometers. As a result, a velocity dispersion function related to varying wave heights are derived with a statistical approach for vegetated and non-vegetated environments. Multiple regression method is used in the statistical approach and the aforementioned functions are given for the parameters which are relative temperature, dimensionless velocity and relative depth. Note that, multiple regression method was previously used for obtaining the dispersion coefficient and the flow conditions. The thermal water gradient affected by vegetation with respect to the mentioned variables was tried to be determined in the flow field.Excess temperatures for the vegetated experiments were chosen as 5 and 15 oC, where the flow rate was 15 l/min and the generated waves in the flume were 2.5, 5 and 7 cm of height for each temperature value. The experiments of non-vegetated case were carried out with the flow rate values; namely, 10 and 15 l/min for the same wave conditions. Recorded velocity values were divided by wave group velocity (H/T). Multiple regression analyses conducted for velocity are valid for the third and the fourth zone, if the temperatures of the first and the second zone are assumed to be constant. Consequently for the vegetated case experiments, the equation U/(H/T) =3.4457*z/zo - 0.00004*L - 0.0066 DT - 0.0329 Q was obtained as a result of the multiple regression analysis carried out for U/(H/T). And the result of the multiple regression analysis carried out for the non-vegetated case comes out with the equation U/(H/T) =2.7662*z/zo - 0.0001*L - 0.0080*DT - 0.0181*Q. The difference between these two equations reveals the change in the flow affected by the vegetation. This difference may be computed by a subtraction operation between these two equations which leads to the equation [U/(H/T)]s-[ U/(H/T)]= 0.6795 z/zo+ 0.00006*L + 0.0014*DT- 0.0148*Q. Here the index S expresses the result of the vegetated case experiment. The variables of the newly given equation come out as relative depth, horizontal distance, temperature difference between the receiving environment and the discharged fluid, and the flow rate of the discharge, all of which should already be known by any engineer who works on subjects related to thermal water discharge. Therefore, determination of the change in temperature and the velocity gradient becomes a simple issue. Note that, wave steepness and wave period are on the right hand side of the recommended equation (wave steepness is obtained by dividing significant wave height by significant wave period as known), and it is clear that if one knows the design wave height and wave period, it becomes possible to compute the vegetated case velocities after the determination of non-vegetated case velocities with the help of the given equations. Obtained results are valid for the scope and conditions of this study, and more general approaches should be produced by future studies on this issue. Specifically, investigating the effect of variation of the vegetation properties on the mentioned phenomenon can be recommended. Keywords: Thermal discharge, vegetation, velocity variation.
Data sources:
Access Repository Dashboard
Do you manage itüdergisi (E-Journals - İstanbul Teknik Üniversitesi, Istanbul Teknik Uniersity., İTÜ)? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!