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
Effect of dual-permeability structure in a porous media on heat and mass transfer by natural convection
Get PDFBu çalışmanın amacı, içerisinde ısı ve kütle üretimi olan ve geçirgenliğin tektürel olmadığı gözenekli ortamda doğal taşınımla ısı ve kütle geçişini Darcy-Brinkmam modeli kullanarak incelemektir. İki boyutlu akışkana doymuş gözenekli ortamdaki akış için Soret ve Dufour etkisi ihmal edilmiş ve yoğunluk için Boussinesq yaklaşımı kullanılmıştır. Elde edilen boyutsuz korunum denklemleri hücre merkezli sonlu hacimler yöntemi ile ayrıklaştırılmıştır. Boyutsuz akım fonksiyonları “Succesive Over Relaxation”; boyutsuz derişiklik, enerji ve çevrinti taşınım denklemleri “Alternating Direction Implicit” yöntemleri ile çözülmüştür. Farklı geçirgenliğe sahip bölgelerin gözenekli ortamdaki ısı ve kütle geçişine etkisini incelemek için iki farklı geçirgenli yapıyı temsil edecek altı farklı model oluşturulmuştur. Bu modellerin geçirgenlik değerleri 10-7ve 10-4 olmak üzere iki farklı Darcy sayısı (Da) ile temsil edilmiştir. Yapılan sayısal çalışmalarda gözeneklilik (ε=0.4), Prandtl sayısı (Pr=0.7), boyut oranı (A=1), Lewis sayısı (Le=10), yüzdürme oranı (N=2) sabit tutulmuştur. Farklı Rayleigh sayıları (Ra) kullanılarak yapılan incelemenin sonuçları, akım fonksiyonu, eşderişiklik ve eşsıcaklık eğrileri kullanılarak sunulmuştur. Sonuçlar, ayrıca Nusselt (Nu), Sherwood (Sh) sayılarının ve normalize edilmiş derişiklik değerlerinin değişimi dikkate alınarak değerlendirilmiştir. Çalışmada, ikili geçirgenli yapı ile oluşturulan modellerin ısı ve kütle geçişi üzerindeki etkilerinin yüksek Rayleigh sayıları için farklılaştığı sonucuna varılmış. Ayrıca, düşey eksen boyunca oluşturulan ikili geçirgenli yapı, yatay eksen boyunca oluşturulan ikili geçirgenli yapıdan daha etkili olmuştur. Bunun sebebi geçirgen kısmın tamamının yüksek Darcy sayısına sahip olmasıdır. Anahtar Kelimeler: Doğal taşınım, gözenekli ortam, Darcy-Brinkman modeli, geçirgenlik, Darcy sayısı, Rayleigh sayısı.The main objective of this work is to investigate the effect of the structure of double permeability on double diffusive natural convection. The permeability, K, is the measure of the flow conductance of the porous medium. In this study, two main cases are taken into account. In the first main case, the cavity has homogenous permability in everywhere and in the other main case; the cavity has two regions with different permeability values. The permeability of the porous medium can be represented by Darcy number, since Darcy number depends on the permeability, directly (Da=K/L2). Therefore, Darcy number is used instead of the permeability for all cases. The natural convective flow with heat and mass transfer for porous media, saturated with two dimensional fluid, having heat and mass production in solid phase was investigated by using Darcy-Brinkman model. The fluid is assumed to be a normal Boussinesq fluid that is the density variations upon temperature and concentration at constant pressure. Soret and Dufour effects on heat and mass diffusion were neglected. The cavity is cooled from all the walls, where as the right wall is partially permeable and other boundaries are impermeable. The double-diffusive natural convection in a porous cavity with partially permeable wall is analyzed by solving mass, momentum, vorticity, energy and concentration balance equations, using the Brinkman extension of the classical Darcy equation. A cell-centered finite volume scheme is applied to solve the governing equations. The stream function field is calculated by using a Successive Over Relaxation method (SOR). The grid layout was arranged by utilizing collocated grid procedure, while the power law-differencing scheme was adopted for heat, mass fluxes and vorticity transfer in the fluid domain. The iterative procedure is performed with the Alternating Direction Implicit method (ADI). For benchmarking purpose, the accuracy of the numerical code was checked in the case of double diffusive convection within a differentially heating square porous enclosure subject to a concentration difference, C, and a temperature difference, T in the horizontal direction using the results reported in Goyeau et al. (1996). The results concern mass transfer due to purely thermal natural convection (N=0) for Darcy model (Da=10-7). When the Darcy number is small enough (Da 10-7) the numerical results obtained with the Brinkman model are in agreement with Darcy's law (Bennacer et al., 2001). The average Nu and Sh numbers are in good agreement with those published in the references. The grid was selected as a trade off between numerical accuracy, stability, and computational time. A non-uniform grid (64x64) was used in all calculations. Heat and mass transfer characteristics as isoconcentration lines, streamlines, isotherms, average Sherwood numbers and Nusselt numbers were studied for different values of Rayleigh number and the structures of double permeability. The main findings of this investigation are as follow: The concentration gradients in all cases are larger according to the medium with (Da 10-7) homogenous permeability. On the other hand, the temperature gradients in all cases are smaller according to the medium with (Da 10-7) homogenous permeability. In case 3 and case 6, the Darcy number of the medium is 10-7 in 75 %. Therefore heat and mass transfer characteristics as isoconcentration lines, streamlines, isotherms, average Sherwood numbers and Nusselt numbers were expected to be similar, but not seen any similarities between case 3 and case 6. There are similarities only between case 6 and case 0 for the concentration gradients and the average Sherwood numbers. The average Nusselt numbers for six cases are nearly similar. Also, the average Nusselt number increases as the Rayleigh number increases for all cases. The maximum changing of Nusselt number is seen at case 5, the minimum changing is case 0. As the Rayleigh number increases, the amount of the leakage from the cavity also increases. As a result, the leakage from the cavity has higher values in the double permeability cases that the Darcy number of the bottom region of the cavity is 10-7 in 75 % (case 6). Keywords: Natural convection, porous medium, permeability, Darcy number, Rayleigh number, Darcy-Brinkman model.
Researching the effects of the expertise levels of the participants to the results on national technology foresight Delphi Survey
Get PDFTeknoloji öngörüsü, bilim, teknoloji, ekonomi, çevre ve toplum geleceği ile ilgili uzun dönemli beklentilerin dikkate alınması ile ülkeye gelecekte en fazla ekonomik, çevresel ve sosyal yararlar sağlayacak yayılgan teknolojilerin ve stratejik araştırma alanlarının belirlenmesi için yapılan sistematik çalışmalar bütünüdür. Öngörü, bugünkü seçimlerin geleceği şekillendirebileceği hatta yaratabileceği olgusuyla, geleceğe yönelik bilinçli bir aktif yaklaşımdır. Konu yaygınlığı ve önemi itibariyle çok sayıda bilim insanının ilgi alanı olmakta, literatürde metodolijelere ilişkin tartışmalar ön plana çıkmaktadır. Son dönemde üzerinde özellikle çalışılan alanlardan birisi de, görüşüne başvurulan ‘uzmanları’ın; -interdisiplinerlik yaklaşımını gözeterek- uzmanlık seviyesi, temsil ettiği sektör, sosyo-ekonomik profili, cinsiyeti, yaşı gibi faktörler dikkate alarak seçilmesi konusudur. Çalışma kapsamında uzmanlık konusuna, bilimsel yaklaşım çerçevesinde bakılmış, Ulusal Öngörü Projesi Vizyon 2023 Delfi Analizi verilerinden TÜBİTAK izni ile yararlanılarak, görüşüne başvurulan uzmanların, uzmanlık seviyesinin çıktılara etkisi, istatistiki olarak analiz edilmiştir. Ortaya konan hipotez; ulusal teknoloji öngörüsü projelerinde yaygın danışma yöntemi olarak kullanılan Delfi Analizi’ne katılan ve görüşünden yararlanılan uzmanların, uzmanlık seviyelerinin bir faktör olarak hesaba katılmasının, katılmaması durumu ile fark arz ettiği, dolayısıyla çalışma sonucunda belirlenen öncelikli teknoloji alanları sıralamasında farklılık yaratacağıdır. Yapılan istatistiki değerlendirme ve uygulanan testlerle hipotez doğrulanmıştır. Ulusal teknoloji öngörü projeleri geleceği kendi kurgulayabilen bir güç olma yetkinliğini kazanmak isteyen ülkeler için kritik öneme haizdir. Bu alanda; metodolojiye, uygulama ve sonuçlara yönelik yapılan araştırmaların, veriler analiz edilerek ulaşılan yeni sonuçların, sonraki çalışmalar için temel teşkil etmesi ve gelişme fırsatları sağlaması hedeflenmektedir. Anahtar Kelimeler: Ulusal teknoloji öngörüsü, Delfi Analizi, uzmanlık seviyesi.Foresight is the whole systematic studies performed for determining spreading technologies and strategic researches which will provide the most economic, environmental and social benefits to the country in the future by taking into account of the long term expectancies relating to science, technology, economy, environment and society's future. In its other definition, it is a systematic process which includes to concern with science, technology, economy and society's long term perspective in order to define strategic researches and spreading technology fields. Foresight studies can be gathered under three generation. In the first generation studies, foresight consisted of only technology forecasting. Technology experts and futurologists were participated in these foresight studies, in the second generation foresight studies, market and technology were considered together and participants consisted of academicians and industrialists. And in the third generation foresight studies, social dimension was accompanied to technology and market.. Expertise levels of experts who are consulted for their opinions and their effects to the results of foresight are subjects of the discussion in this extent. The subject is concerned within the frame of scientific approach by basing on these discussions and the effects of expertise levels of the experts consulted to the outputs is measured statistically by making use of data with the permission of TÜBİTAK from Vision 2023 delphi survey, the first national technology foresight project of Turkey. The subject on selecting experts consulted for their opinions delphi surveys according to their expertise levels, which is discussed many countries and literatures within the context of National Technology Foresight projects is studied. The matter that the persons having knowledge and experience regarding the subject intensively being in agreement recently in the literature is in the direction that every expertise level adds a different point of view to the study. The argument that the studies including only the experts regarding the subject reflect a narrow point of view is common However, in the articles written before 2000, traditional approach defense was in the direction that studies only with experts having higher and medium expertise levels provides to focus on more specific technology fields and obtains more effective results scientifically. Japan Delphi survey still carries out its performances with experts having higher expertise levels. “Expertise” subject taking place in the literature is examined with a new agenda by approaching more different point of view within the study extent. The effect of taking into account of determination of expertise levels and specific/strategic technology fields of the expert consulted both in panels and in Delphi survey as a factor to the results is examined statistically within the context of national technology foresight project. Hypothesis is that taking into account of expertise levels of the experts participating in Delphi Survey and consulted for their opinions as a factor creates a difference in the grading strategic technology fields determined as a result of the study. Data of Delphi Survey is firstly evaluated by considering the expertise levels of all experts equal and then new results are occurred by weighting the answers they stated for every statement with their expertise levels. Two different results from the answers the same sample group gave are compared to each other and are entered to appropriate tests. The results of these tests prove that the contribution values are different from each other in cases of the same expertise level and of the different expertise level of all participants. And this shows that taking into account of expertise level as a factor causes differentiation of the strategic technology fields determined according to Delphi Survey performed within the context of foresight study and confirms hypothesis. National technology foresight projects have a critical importance for countries wishing to gain competency to be a power for building the future on their own. In this field; it is expected that new results obtained with scientific approach by analyzing the researches made for methodology, performance and result, evidences put forward with different approaches will be a base for next studies and provide important improvement opportunities. Keywords: National technology foresight, Delphi analysis, expertise level. 
Critical control times in optimal control of linear quantum system
Get PDFBu çalışmada doğrusal yapılandırılmış kuvantum sistemlerin eniyilemeli denetimi üzerine odaklanılmıştır. Hareket denklemlerinin oluşturumu ve sonra da çözümü üzerinde çalışmalarda bulunulmuştur. Potansiyel tanımında konuma bağımlılıkta, bağımsız değişkenlerin en çok ikinci dereceden olan kuvvetlerinin ya da onların ikili çarpımlarının doğrusal birleşimi alınmaktadır. Bu yapılar, sonunda birinci basamaktan sıradan türevli bir denklem takımıyla tanımlanabilen duruma getirilebilmektedir. Bu denklem takımında bağımsız değişken zaman parametresi olmakta ve değeri 0 ile etkileşme süresini gösteren T değeri arasında değişebilmektedir. Denklem takımına eşlik eden koşulların yarısı etkileşimin başında diğer yarısı ise sonunda verildiğinden “Zamanda Sınır Değer Problemi” nitelikli matematiksel bir yapıyla karşılaşılmaktadır. Çalışma dalga fonksiyonu ile eşdüzey fonksiyonunu değil, özellikle, dış alan genliği ile sapma parametresinin belirlenmesini amaçlamaktadır. Denklemlerin yapılarının elvermesinden yararlanarak, sadece zamana bağımlı olan ve aslında herbiri ya bir beklenen değer ya da dalga fonksiyonunun tanımladığı düzey ile eşdüzey fonksiyonunun tanımladığı düzey arasında geçiş değeri olan bilinmeyenler üzerinde sıradan türevli denklem takımları oluşturmak amaçlanmış ve başarılmıştır. Denklem takımının oluşturulması bütünüyle özgün olarak üretilmiş bulunmaktadır. İncelenen durumlardaki doğrusal yapılandırım, elde edilen sıradan türevli denklemlerin de doğrusal yapılı olmasına olanak sağlamıştır. Sınır değerli bu denklemlerin, ağırlıkların analitik olması durumunda kesin olarak çözümü sağlanmış diğer durumlar için önerilen bir saptırım açılımınınsa tüm parametre değerleri için yakınsak olacağı gösterilmiştir. Eniyilemeli denetimde denetim süresinin çözümün gerçekte uygulanabilirliğini çok önemli biçimde etkilediği gösterilmiştir. Sayıları sonsuz olabilen kritik denetleme süreleri için dış alan genliğinin ve sapmanın sonsuza gidebildiği bazı denetim sürelerinin ise sapmayı sıfırlama olanağı verebildiği de gösterilmiştir. Anahtar Kelimeler: Kuvantum mekaniği, optimal kontrol, dalga fonksiyonu, harmonik salınıcı.This work focuses on the evaluation of critical control times in the optimal control of linear quantum systems. The linearity is used to imply the existence of only linear intra forces in the system. This means that the system under consideration is composed of harmonic oscillators. The linearity is not only in the forces. The external field's dipole polarizability function is also taken linear in spatial variables. Beyond that the objective operator whose expectation value's deviation from its target value should be minimized in square and the penalty operator whose expectation value is to be suppressed are also linear in position and momentum operators. We assume that the system has 3N degree of freedom and the cost functional for the optimal control is composed of four terms: objective term which is the square of the objective operator whose expectation value's deviation from its prescribed target; penalty term for getting finiteness in external field; penalty term for an operator whose norm is to be suppressed ; dynamical constraint term which enters the quantum motion of the system to the optimisation. This cost functional depends on five unkowns: (1-2) wave function and its complex conjugate; (3-4) costate function which is in fact the Lagrange multiplier of the dynamical constraint term. It varies in time and space and somehow describes the backward motion of the system from the final instant of the control. It is a complex valued function hence its complex conjugate is the companion unknown to itself; The external field amplitude within the dipole the polarizability approximation. It is temporal and real valued. There may be a shorthand notation for the deviation of the expectation value of the objective operator from its target value. In that case, this entity can be considered as another unknown and can be handled by adding its definition as an extra condition. The equations of motion can be optained by setting the first variation of the cost functional to zero. Four partial differential equations are obtained for themselves and complex conjugates of the wave and costate functions. The equations for the wave function are accompanied by initial conditions while the costate function related partial differential equations are right hand sided and accompanied by final conditions. This for equations two by two, describes forward and backward evolutions for the system and these evolutions are connected by the algebraic however functional relation obtained for the external field amplitude. All these equations are non linear for the general case. However for the present linearity assumptions they become linear. The equations of motion mentioned above are not attempted to be solved. Instead, ordinary differential equations for the expectation values of position and momentum operators on the wave function and the transition terms between the states decscribed by the wave and costate functions are constructed. These temporal ODEs are linear but may be timevariant in the coefficients if the weight functions of the penalty terms are temporally varying. The simplest case involves the cases where these weights are constansts and the ODEs can be analytically solved in terms of an exponentials matrix. This results in analytic expressions for the unknowns. These expressions depend not only on time but olsa control time (T). As can be shown by rigorous analytical analysis and verified by the numerical implementations, there are certain critical values of the control time where the extressions grow up to infinity or vanishes. Vanishing values mean exact achievement of the target value by the expectation value of the objective operator. These are the best control instants where the goal is exactly achieved. On the other hand, control times causing infinite jumps, are just the time instants the control becomes infeasible. The critical control times creating infinite jumps are called "Critical Control Times for Uncontrollability" and it is important to know these values before attempting the experimentation to avoid uncontrollability. Otherwise, experiment is wasted out. The critical control times which make the deviation of the expectation values of the objective operator from its target value are called "Critical Control Times for Exact Achievement" and they are useful to get the best control. Paper contains all important issues about these points. Keywords: Quantum mechanics, optimal control, wave function, harmonic oscillator
Subtractive Approach to fuzzy c-means clustering method
Get PDFGörüntü işleme, uzaktan algılama, veri madenciliği, örüntü tanıma ve benzeri konularda yaygın olarak kullanılan kümeleme yöntemleri, bir grup içindeki benzerliklerin gruplar arasındaki benzerliklerden daha büyük olmasını amaçlamaktadır. Farklı yoğunluklara sahip kümeler içeren veri uzayları için kümeleme işlemi zordur ve bu problemi çözmeye odaklanan birçok çalışma ileri sürülmüştür. K-means ve bulanık c-means kümeleme yöntemlerinin performansı küme merkezlerinin başlangıç değerlerine bağlıdır. Bu yüzden her iki algoritmanın da farklı küme merkezi başlangıç değeri için birçok defa çalıştırılması gerekir. Çıkarımlı kümeleme yöntemi ise veri noktalarının konumlarından veri uzayının yoğun bölgelerini tespit etmeye ve en çok komşuluğa sahip olan veri noktalarını küme merkezi olarak seçmeye dayanır. Bu özelliğiyle başlangıç koşulundan bağımsızdır ve algoritmanın bir kez çalıştırılması yeterlidir. Ancak, küme merkezleri veri noktalarından başka konumlardan saptanamadığı için bu yöntem her veri uzayına uygun olmayabilir. Bu makalede önerilen kümeleme yöntemi sayesinde genel kümeleme yöntemlerindeki başlangıç koşulu, ayrıca çıkarımlı kümeleme yönteminde küme merkezlerinin veri noktalarından seçilme zorunluluğu ortadan kaldırılmıştır. Dört yapay veri uzayı ile test edilen yeni yöntem, k-means, bulanık c-means ve çıkarımlı kümeleme yöntemleri ile karşılaştırılmıştır. Sonuç olarak bulanık c-means ve çıkarımlı kümeleme yöntemlerinin avantajlarını birleştiren yeni yöntem ile bulanık c-means yönteminin başlangıç koşuluna bağımlılığı ve küme merkezlerinin veri noktalarından seçilmesi zorunluluğu ortadan kaldırılmıştır. Anahtar Kelimeler: başlangıç koşulu, k-means, bulanık c-means, çıkarımlı kümeleme. Data clustering is an important part of cluster analysis. Based on various theories, numerous clustering algorithms have been developed, and new algorithms continue to appear in the literature. The aim of clustering is to obtain the most different groups in a dataset. A clustering method finds the similar data points and puts them into groups. If the groups in a dataset are found, then the dataset can be represented by fewer symbols. In the literature, researchers have proposed many solutions for this issue based on different theories. But there are still some problems such as the optimum cluster centers and the initial condition. The best-known and earliest clustering method is K-means clustering algorithm. Its main advantageous is the capacity of fast converging. In spite of having many successful applications in several fields, it has many drawbacks. The membership values being only 0 or 1 may not always reflect the practical relationship between the data point and the cluster. In order to cope with this drawback, fuzzy c-means method employs fuzzy partitioning so that each data point can belong to several clusters with membership values between 0 and 1. Both clustering techniques try to group the data into given the number of clusters. Another method, subtractive clustering, finds the largest cluster by using the density function, then the second one, and so on. Subtractive clustering method uses the locations of the data points to calculate the density function. K-means method tends to making homogenous distribution. Fuzzy c-means clustering method makes clusters with soft edges. Subtractive clustering usually tries to find the discreteness. The locations of cluster centers in K-means and fuzzy c-means clustering may not be same for each time because of depending on initial condition. Therefore, they should be run several times for all datasets. Subtractive clustering method has only one solution independent of initial condition; consequently, it is enough to run once. But the main problem of subtractive clustering method is that the cluster centers are selected among data points. Because the cluster centers selected among data points may not represent the clusters of dataset. In this paper, we offer a new approach which combines fuzzy c-means and subtractive clustering methods. The novel approach takes account of both discreteness and soft edges distribution; so the result has a similar appearance to average of other methods. The three main contributions of new approach can be summarized as: it becomes a more sophisticated technique by taking advantages of fuzzy c-means and subtractive clustering methods; it removes the initial condition. It has also only one solution independent of initial condition as in subtractive clustering method. The novel algorithm consists of the following steps: Step1. Normalize the data points. Step2. Calculate the density value of each data point by Equation (10). Step3. Select the point having the highest density value as cluster center. Step4. Update the densities of each data point by Equation (11). If the number of detected cluster centers is less than the desired number, then go to Step3. Step5. Compute the membership matrix by Equation (7). Step6. Update the cluster centers by Equation (6). Step7. Calculate the cost function by Equation (5). If it is bigger than the selected threshold value, go to Step5. Clustering methods are usually evaluated and tested by using the artificial datasets. These methods must be able to analyze the datasets with different feature and sampling size. Artificial datasets used in the literature have some properties such as symmetric, discrete, and identical form. Therefore, we have used many special datasets in the numeric examples. Finally, the novel approach is successful for both symmetric-identical and asymmetric-non-identical datasets. It also removes dependence on the initial condition in contrast to common KM and FCM clustering methods. Keywords: K-means; fuzzy c-means; subtractive clustering
Atmospheric concentration of Polycylic aromatic hydrocarbons in Istanbul
Get PDFÇalışmada, İstanbul’da ikisi şehir atmosferini (Yıldız, DMO) ve biri kırsal alanı yansıtan (Kilyos) üç örnekleme noktasından, dört mevsimi kapsayacak şekilde, Eylül 2006-Aralık 2007 tarihleri arasında seçilmiş örnekleme periyodunda, alınan numunelerde EPA (Environmental Protection Agency) tarafından öncelikli kirleticiler listesinde bulunan 16 tür PAH (Polisiklik Aromatik Hidrokarbon) ve TSP (Toplam Askıda Katı) analizi yapılmış ve bulunan sonuçlar zamana, mekana bağlı ve kaynak belirlemeye yönelik olarak değerlendirilmiştir. Yıldız örnekleme noktasından 135, DMO’dan 129 ve Kilyos’tan 62 adet olmak üzere toplam 326 adet numunede analiz yapılmıştır. Ortalama PAH konsantrasyonu Yıldız, DMO ve Kilyos istasyonları için sırasıyla 100.66±61.26, 84.63±46.66 ve 25.12±13.20 ng/m3 ve TSP konsantrasyonu 101.16±53.22, 152.31±99.12, 49.84±18.47 µg/ m3 olarak ölçülmüştür. Tüm istasyonlardan alınan sonuçlar arasında, en yüksek toplam PAH değerleri Yıldız örnekleme noktası için ve en yüksek TSP değerleri DMO istasyonu için tespit edilmiştir. En düşük konsantrasyonlar Kilyos istasyonu için belirlenmiştir. Tüm istasyonlarda PAH türleri arasında konsantrasyonu en yüksek değerlerde gözlenenler molekül ağırlığı düşük PAH’lar olmuştur. Şehir atmosferini temsil eden alanlarda güçlü bir mevsimsel değişim gözlenmiştir. En yüksek PAH konsantrasyonları kış aylarında ve şehir atmosferinde gözlenmiş en düşük konsantrasyonlara ise yaz aylarında ve kırsal alanda rastlanmıştır. PAH’ların kaynaklarını belirlemek amacıyla “kaynak tanımlama katsayıları” ile yapılan belirlemede Yıldız ve DMO istasyonları için trafik emisyonunun ağırlıklı olduğu ve tüm belirleme oranları için dizel emisyonlarının benzinli araç emisyonlarına oranla baskın olduğu gözlenmiştir. Anahtar Kelimeler: Polisiklik Aromatik Hidrokarbonlar (PAH)), Toplam Askıda Katı (TSP), kaynak belirleme katsayıları.Polycyclic aromatic hydrocarbons (PAHs) are semivolatile organic compounds consisting of two or more fused aromatic rings. These compounds are produced in the atmosphere as by-products of the incomplete combustion of almost any fuel. After emitting the atmosphere, this pollutants fallout to the ground as dry or wet deposition depends on its form and many factors. The concentration of PAH in the atmosphere is in the order of ng/m3. Atmospheric PAHs are partitioned between the particulate and the gaseous phases. PAHs have received increased attention in recent years in air pollution studies because some of these compounds are highly carcinogenic or mutagenic. Some PAHs were classified as probable or possible carcinogens to humans. For this reason, especially after 1980?s years, concentrations of these pollutants have been monitored continuously, especially in water mass and atmosphere. In many metropolitans, concentrations and potential toxicity of PAHs to city ecosystem have been determined in monitoring network or individual studies. In the base of these monitoring programs, sources of PAHs identified and quantified at receptor locations with different techniques. This study is the first about atmospheric concentration of PAHs in Istanbul. In this study, 16 EPA`s PAHs and TSP (Total Suspended Solids) concentration were performed in three sampling stations for four seasons in the period of September 2006-December 2007 in Istanbul. Spatial and temporal profiles were determined and diagnostic ratios were applied for source apportiontment of the pollutants. Airborne samples were collected using PS-1 samplers with GFF filter to collect airborne particle-bound PAH and TSP. The filter was followed by a glass cartridge containing polyurethane foam plug and XAD-2 resin. All sampled filters and glass cartridges were extracted by soxhlet method with n-hexane:diethylether mixture for 20 h. The extract was then concentrated and cleaned up with silica gel column using procedures by EPA. Analysis of all extracts were performed by means of reversed-phase HPLC using UV-visible and fluorescence detectors in series. A total of 326 airborne samples were collected and analyzed for 16 PAHs and TSP in the period at three monitoring stations; Yildiz (135 samples-urban site), DMO (129 samples-urban site) and Kilyos (62 samples-rural site). Total average PAH concentrations were 100.66±61.26, 84.63±46.66 and 25.12±13.34 ng m-3 and TSP concentrations were 101.16±53.22, 152.31±99.12, 49.84±18.58 µg m-3 for Yildiz, DMO and Kilyos stations respectively. When comparing average concentration, maximum PAH values were observed for Yildiz station but maximum TSP values were observed for DMO station. At all the sites, the lighter compounds (MW 200) were the most abundant species. Nevertheless some reduction was observed in percentage of the low molecular weight PAHs during the warm season for all stations. Because PAHs with high molecular weight are found mainly adsorbed in particulate matter, a good correlation between TSP and heavier PAH concentration is expected. In general, correlation values between TSP and total heavier PAH values were good for Yildiz and DMO stations. Average correlation values were 0.60 and 0.56 for Yildiz and DMO stations respectively. The pattern of PAH and TSP concentrations showed spatial and temporal variations. Strong seasonality was observed especially for urban stations. The highest level was observed in winter in urban areas while the lowest in summer in rural area. When comparing seasonal values, Winter 2006 had the highest PAH levels for all stations and the lowest levels were observed in Spring and Summer 2007. Comparing yearly variations, year 2006 had bigger PAH values than year 2007. PAH concentration values were compared literature studies. The comparison of the results with the examples given in the literature was in a good agreement. The concentrations or ratios of specific PAH compounds have been used to indicate towards the corresponding emission sources, especially for vehicle emissions In the study, PAH concentration values were analyzed with diagnostic ratios. The ratios of PAHs were compared with individual concentration values in order to determine sources of PAHs. Yildiz and DMO sites was affected by traffic-related sources, and diesel emission sources had higher contribution than gasoline ones. On the other hand, low level traffic emission was obtained for Kilyos station as expected. Keywords: Polycyclic aromatic hydrocarbons, total suspended solids, diagnostic ratios
Geometric accuracy assessment of stereo SPOT-5 HRG level 1A images using parametric approach considering satellite orbital parameters
Get PDFSon yirmi yılda doğrusal dizi algılama teknolojisi sayesinde hava ve uzay bazlı görüntüleme sistemleri önemli ölçüde gelişmiş ve yer örnekleme aralığı 40 cm'ye kadar küçülmüştür. Böylece görüntüler yardımıyla konumsal uygulamaların gelişimi ve elde edilen konum doğruluğu da artmıştır. Bu makalede, öncelikle doğrusal dizi görüntüleme teknolojisi ve geometrik hataları hakkında kısaca bilgi verilmiş; sonra görüntü ve nesne (yer) koordinat sistemleri arasındaki geometrik ilişkiyi tanımlayan genel bir model tanımlanmıştır. Kullanılan görüntünün özellikleri göz önüne alınarak bu model değiştirilmiş ve basitleştirilmiştir. Dengeleme işlemi bilinmeyenli koşullu ölçüler dengelemesi ile ve ön ve demet dengeleme şeklinde iki aşamalı olarak gerçekleştirilmiştir. Bakış açılarının ve yörünge parametrelerinin doğruluğa etkisi analiz bölümünü oluşturmaktadır. Araştırmada, Zonguldak ve çevresini kapsayan 5 m yer örnekleme aralığına sahip stereo SPOT-5 HRG düzey 1A görüntüleri kullanılmaktadır. Noktaların yer koordinatları bağıl statik GPS yöntemiyle ölçülmüştür. Tüm hesaplamalar ve grafik gösterimler MATLAB ortamında geliştirilen GeoSpot-1.0 yazılımıyla gerçekleştirilmiştir. Görüntülerin analizi sonucunda yüksek geometrik doğruluğa ulaşmada en önemli etkenin iç yöneltmeyi tanımlayan bakış açıları olduğu görülmüştür. Dış yöneltme elemanları sabit kabul edilerek bakış açıları her bir görüntü için ayrı ayrı düzeltilmiştir. Doğruluk değerleri kontrol noktalarında ±1 m'den daha iyi ve denetim noktalarında yaklaşık ±5 m düzeyinde bulunmuştur. Dış yöneltme elemanlarının doğruluğa etkisi iç yöneltme elemanlarına oranla önemsenmeyecek düzeydedir. Ancak dış yöneltme elemanları arası olası korelasyon doğruluğu oldukça düşürmektedir. Analiz, farklı dış yöneltme elemanı grupları ve nokta dağılımları için gerçekleştirilmiştir. Anahtar Kelimeler: Geometrik analiz, geometrik doğruluk, parametrik model, uydu yörünge bilgileri, SPOT-5, HRG, düzey 1 A, koşullu ölçüler dengelemesi, demet dengeleme.In the last two decades the imaging technology in aerial and space based missions grew up thanks to especially the imaging technology based on linear array sensors and reducing ground sampling distance. Now, high resolution space images with about 40 centimeter resolution are available. This improvement supports developments on geospatial applications of these images. Thus, the geometric accuracy increases thanks to the geometric resolution. This improvement causes that geometric correction process becomes more important now than in the past for this purpose. This article is focused on the geometric analysis of high resolution space images using parametric (rigorous) approaches ignoring non-parametric (deterministic) ones. Parametric approaches consider imaging geometry, orbital and attitude parameters of satellite, and define the real geometry between image and ground on contrary to the non-parametric ones. However parametric models are more complex and need many auxiliary coordinate systems defined between image and ground coordinate systems. The analyzed images are only one set of stereo SPOT-5 HRG level 1A images acquired by linear array sensors in 5 m ground sampling distance, because the required orbital information is available only with this image on hand. So this technology, linear array sensors, and its distortions are subjected. The generic model which establishes the geometric relationship between image and ground coordinate systems is defined at first (Weser et al., 2008). The required auxiliary coordinate systems between image and ground coordinate systems, i.e. scanline, camera, payload and orbital coordinate systems are defined. Then the modification and simplification to generic model are explained taking into account the characteristics of image used in this study (SPOT Image, 2002; Fotev et al., 2005). The modification is based on the interior orientation via defining the look angles for each element of CCD line. These look angles are the major dominant component on the geometric accuracy. And the simplification is based on the rotation from payload coordinate system to orbit coordinate system performed by the attitude angles. The ground coordinates and their accuracies are attained by adjustment process requiring both pre- and bundle adjustment stages. The adjustment is performed by the Gauss-Helmert model. The test field covers mainly Zonguldak and partly its neighbors, Bartın, Bolu and Karabük located in the Western Karadeniz Region of Turkey. The test field called Zonguldak test site comes into prominence with its mountainous topography. The mountainous areas are very important to analysis geometric accuracy of images since many images and the mathematical models give very accurate results on the flat areas on the contrary the mountainous areas. Specifications of SPOT-5 HRG level 1A image and auxiliary data are briefly presented. All computations were performed programming in the MATLAB environment developing GeoSpot-1.0. Many point sets are established including different distribution of ground control points (GCPs) and independent check points (ICPs). The GCPs are needed for the adjustment of look angles and the parameters where the ICPs are required for the checking of adjustment. 47 points in total are measured by the static GPS observation by the team from Department of Geodesy and Photogrammetry Engineering in Zonguldak Karaelmas University. The points are obtained considering their homoge-nous distribution both on planimetry and height. The most important issue for the purpose of estimating the true ground coordinated via the stereo images is adjusting the interior orientation components, i.e. look angles to be corrected in the pre-adjustment process, in this thesis. However the effects of exterior orientation parameters on the accuracy evaluation have to be investigated estab-lishing various sets of them. The adjustment requires selection of suitable set of these parameters. The results of geometric analysis are presented with the help of graphical figures and the tables at the end of this thesis. The geometric accuracy achieved is about ±1 m (±0.2 pixel) for the GCPs and ±5m (±1 pixel) for the ICPs. The accuracies are in trend that mZ > mX > mY. This can be expected since the accuracy in Z direction is generally less than the planimetry (X and Y). The accuracy in Y axis is better than in X since the Y is related to the look angles which are the most dominant component on the accuracy. The accuracy of ICPs can be acceptable considering 1 pixel is the key unit of an image. Keywords: Geometric analysis, geometric accuracy, parametric model, satellite orbital parameter, SPOT-5, HRG, level 1A, Gauss-Helmert model, bundle adjustment
A research on the applicability of pricing in Istanbul: Eminonu pricing model
Get PDFTrafik tıkanıklığı modern kent yaşantısının önemli sorunlarından biridir. Özellikle büyük metropollerde, önemli ulaştırma yatırımları yapılarak bu soruna çözüm bulunmaya çalışılmaktadır. Trafik tıkanıklığı sorunun çözümünde yatırımlar kadar etkili olan uygulamalar arasında, ulaşım talebinin parasal veya parasal olmayan yöntemlerle sınırlanması da yer almaktadır. Bu yöntemlerin en önemlilerinden birisi tıkanıklık fiyatlandırmasıdır. Bu uygulama, Londra ve Stockholm kentlerindeki başarılı örnekler ile beraber son yıllarda dünyadaki popülaritesini arttırmaktadır. Tıkanıklık fiyatlandırması, ciddi trafik tıkanıklığı sorunları yaşanan İstanbul için de uygun bir yöntem olmasına rağmen, yoğun nüfus ve plansız yapılaşmanın sonucu olarak ortaya çıkan karmaşık ulaşım ağının, uygulanabilirlik açısından çeşitli güçlükler yaratması söz konusudur. Bu çalışmada, tıkanıklık fiyatlandırmasının İstanbul’da uygulanabilirliğinin araştırılması amacıyla, eski Eminönü ilçesinde gerçekleştirilecek bir kordon fiyatlandırması uygulaması için bir fiyatlandırma modeli geliştirilmiştir. Eminönü, ev sahipliği yaptığı tarihi ve kültürel varlıklar ile kentin önemli bölgelerinden biridir. Bölgenin küçük yüzölçümü ve düşük yerleşik nüfusuna karşılık sürekli ziyaret edilen bir yer olması, bölgede ciddi trafik tıkanıklığı ve otopark sıkıntıları ortaya çıkarmaktadır. Fiyatlandırma modeli, sosyal optimumunu sağlamak için uygulanan optimum fiyatın yanı sıra, farklı fiyatlar için de değerlendirmeler yapmaya olanak veren bir modeldir. Modelden elde edilen sonuçlar, iyi planlanmış bir fiyatlandırma uygulaması ile yalnızca trafik tıkanıklığı sorununa çözüm üretilmeyeceği, ciddi düzeyde ilave maddi kaynak da yaratılacağını ortaya koymaktadır. Anahtar Kelimeler: Trafik tıkanıklığı, talep yönetimi, tıkanıklık fiyatlandırması.Congestion is one of the major problems of metropolitan cities. This problem arises when the increase in demand for travelling with private vehicles surpasses the capacity of the road network. Travel demand management strategies are developed to restrict the excessive demand without making expensive infrastructure investments. Demand management strategies can be classified in two main groups: Fiscal and non-fiscal strategies. Fiscal strategies are implemented to regulate the demand by arranging taxes or tariffs collected for the usage of the transportation infrastructure or facilities. Non-fiscal strategies aim to adjust the demand for private vehicle usage without imposing any financial measures on individuals. Congestion pricing is one of the most powerful fiscal strategies. By definition, congestion pricing is collecting fixed or variable amount of tolls from private vehicle users for the usage of a road or a road network. For variable toll implementations, the amount can be determined according to units such as distance covered or congestion level of the facility. Any driver considering entering a congested road will base his/her decision solely on the costs he/she is going to bear. However, entering an already congested road will impose external costs on other drivers currently on the road, as well. The last vehicle entering an already congested road will increase not only the travel times but also the fuel consumption and emissions of all other vehicles. In the case when natural equilibrium occurs, although abovementioned costs already exist, none of the users are paying the price of creating them. In order to make the users pay the price of the external costs they create, marginal cost pricing is implemented. The price collected by marginal cost pricing implementations is named the optimum price and its value is equal to the external costs that users create. Congestion pricing can be modelled by using two different modelling approaches: Time-independent static models and time-dependent dynamic models. Time-independent static models deliberately ignore the hourly changes in the traffic parameters and accept an unchanged relationship between parameters of traffic throughout the analysis. On the contrary, time-dependent dynamic models include the time dimension to the analysis and model the implementation with regard to departure time, as well as route choice. The origins of dynamic models is the bottleneck model which is originally developed by Nobel Prize winner researcher William Vickrey. In this study, a congestion pricing scheme is modelled for Eminonu, in order to evaluate the applicability of congestion pricing to Istanbul. Among all the districts of Istanbul, Eminonu have unique properties. First of all, the district which is a peninsula has clearly defined borders, easing the access control issues. Moreover, due to the small surface area of Eminonu, the road network is relatively small and less complicated. As the host of many historic, recreational and governmental areas, severe congestion is routinely experienced on the road network of the district. The focal point of the modelling studies for Eminonu is the determination of optimum price which is calculated according to the principles of the static models of congestion pricing. Analysis of the equilibrium conditions is made possible by obtaining, average cost, marginal cost and demand curves representing the whole district. In practice, marginal cost pricing may sometimes be ignored or not preferred to be implemented by decision makers. Thus, an evaluation of other pricing regimes different than optimum price is also included in the study. Furthermore, the gains, losses and related measurable economic quantities for the individuals are also calculated. The optimum price is calculated as 0.58 TL/km. It is expected to record a decrease close to 16% in the usage of private vehicles by collecting this toll. Maximum revenue can be generated by collecting a toll of 5.042 TL/km which reduces the number of private vehicle users more than 76%.Futhermore by deploying the average travelling distance of 5.32 km in Eminonu road network, per km prices can be converted into constant prices. Consequently, the optimum price is equal to 3.09 TL in constant values. This study provides valuable information for the applicability of congestion pricing to Istanbul. Obviously, when a congestion pricing scheme is introduced significant amount of individuals will start to travel without their private vehicles and considerable amount of money will be collected In order to secure sustainability and acceptability of the scheme, the revenue collected from the individuals must be returned to the system in favour of users who are worse-off by congestion pricing. Keywords: Traffic congestion, demand management, congestion pricing
Integrating product design and job scheduling on parallel machines
Get PDFBir üretim sisteminde temel amaç, müşteri taleplerinin zaman ve kalite boyutunda etkin bir şekilde karşılanmasıdır. Bu amacı gerçekleştirmek için üretimin büyük partiler halinde ve stok tutarak gerçekleştirilmesi daha mantıklı görünür. Ancak bu yaklaşım, üretim sistemindeki problemleri görünmez hale getirmekle beraber, uzun bekleme sürelerini içermesi nedeniyle üretim temin süresini, dolayısıyla katma değerli olmayan faaliyetlerin oranını arttırmaktadır. Bu durum neticesinde, firmaların hızlı yanıt kabiliyeti ve maliyet performansı olumsuz etkilenir. Paralel makinalarda sıraya bağımlı hazırlık sürelerine sahip işlerin çizelgelenmesini fazla üretim israfına yol açmayacak şekilde sağlayan bu çalışmada, öncelikle, işler arası sıraya bağımlı hazırlık süreleri, ürün tasarım özelliklerinden faydalanan bir matematiksel model ile belirlenmektedir. Böylece, sıraya bağımlı hazırlık süreleri bire-bir ölçüme gerek kalmadan belirlenebilmeleri ve hem güncel hem de güvenilir olmaları sağlanmıştır. Ayrıca, temin sürelerinde önemli bir yer tutan hazırlık sürelerinin düşürülmesi için sıraya bağımlı hazırlık süreleri, odaklanılması önem taşıyan unsurlar arasındadır. Bu bağlamda, sıraya bağımlı hazırlık süreleri toplamını azaltmaya yönelik olarak literatürde yer alan iki sıralama algoritması seçilmiş ve tasarım özelliklerine bağlı bir sıralama algoritması tasarlanmıştır. Sıralama safhasından sonra geliştirilen İş Yükü Bazlı İş Gönderme Sistematiği ile üretim sistem karakteristiğinin çekme sistemine yakınlaştırılması amaçlanmıştır. Geliştirilen metodolojinin performansını değerlendirmek üzere, gerçek bir elektrik donanımı üretim sisteminin kablo kesim-sıyırma-terminalleme sürecine odaklanılmıştır. Belirtilen süreçteki hazırlık işleri analiz edilmiş, ürün tasarım özellikleri belirlenmiştir. Gerçek üretim verileri ile sıralama yöntemlerinin karşılaştırılmasından sonra İş Yükü Bazlı İş Gönderme Sistematiği uygulanmıştır. Anahtar Kelimeler: Çizelgeleme, ürün tasarımı, iş yükü bazlı sipariş yönetimi, paralel makinalar.To gain the ability of designing, scheduling, and manufacturing the products, which customers exactly demand, in time provides competitive advantage to companies. The above mentioned ability is provided by implementing pull system between the costumer and supplier as well as manufacturing processes instead of push system. Such an approach is the reflection of being costumer oriented in production. In addition, setup time constitutes an important part of production lead time. Pull systems lead to significant achievement in terms of lead time when applied in the environment of short setup times which allows production in small batches. Scheduling problems are related to manufacturing resource planning (MRP) in the industrial context. Production managers must select an appropriate scheduling approach that will provide cost-efficient running of the production system. A scheduling approach that guarantees the cost efficient running of the factory is to minimize operational costs that include setup costs (times) which are very important in manufacturing industries. The setup time has often been considered to be negligible or as a part of the processing time. Otherwise, setup times are considered in two types, namely sequence independent and sequence dependent setup times. Sequence independent setup time depends only on the job to be processed while sequence dependent setup time depends on both the job to be processed and the immediately preceding job. Parallel machine problems, where each job can be processed on any of the machines but needs to be processed by only one machine, have a high level of importance in industrial scheduling because of the fact that many examples of the use of parallel machines can be found in the real world. The measurement of sequence dependent setup times becomes inefficient in terms of cost and time in production systems where increased product variey exists. Furthermore, in such systems, it is very difficult to update measured sequence dependent setup times. In this context, it is very efficient to utilize product design characteristics for the estimation of sequence dependent setup times. What is more, it is known in practice that controlling input rate has a great impact on the system performance. In manufacturing systems, this input regulation is performed by the Order Review/Release (ORR) function. Regarded as an effective capacity management tool, ORR manages the transition of production orders from the planning system to the shop floor. In ORR, production orders, that may be generated from a requirements planning system or directly originated from customer orders, are not immediately released to the shop floor without considering any information about the system or job characteristics. These jobs are often first collected in a pre-shop pool and then released to the system according to specific conditions. In this sense, ORR procedures can be considered as the link between production planning and production control as well as the link between manufacturing strategy and day by day management. The purpose of ORR is to improve system performance by controlling the flow of production orders to the system (i.e., the timing and conditions of order release decisions). These improvements can be achieved in terms of increased flexibility, decreased work in process (WIP), improved delivery performance, and decreased congestion and manufacturing lead times. When the relevant literature is reviewed, it is seen that there is not any scientific guiding work which not only considers product design in job scheduling, but also aims to prevent overproduction waste. In this thesis, an integrated methodology that uses product design specifications for job scheduling under the principle of Order Review/Release is developed. The proposed methodology is applied to a real wire harness industrial problem involving the cutting of cables with different colors, sizes and terminals on identical parallel machines with long sequence dependent setup times. The sequence dependent setup times are determined via a mathematical model without measurement. The scheduling objective is to minimize total sequence dependent setup time. Keywords: Scheduling, product design, order review/release, parallel machines
Numerical Simulation of incompressible laminar flow over three dimensional rectangular cavity
Get PDFBu çalışmada basit bir geometriye sahip olmasına rağmen oldukça karmaşık bir akım yapısı sergileyen ve gerek askeri gerekse de endüstriyel pekçok uygulamada sıkça karşılaşılan iki veya üç boyutlu kaviteler üzerindeki zamana bağlı, sıkıştırılamaz akış sayısal olarak incelenmiştir. Bu çalışmanın ana amacı, düşük Reynolds sayılarındaki kavite akışı hakkındaki bilgileri ilerletmek ve 3 boyutlu kavite akışı ile 2 boyutlu (2B) veya 3 boyutlu (3B) kavite üzerindeki akımın 3 boyutluluk etkilerini incelemektir. Sayısal çalışmada, Navier-Stokes denklemleri, zamana bağlı, 3B’lu, sıkıştırılamaz bir akış için Fluent ticari yazılımıyla sonlu hacimler yöntemi kullanılarak çözülmüştür ve türbülans modellemesi için de Large Eddy Simülasyonu (LES) yaklaşımı kullanılmıştır. Çalışmada uzunluğunun derinliğine oranı (L/h) 4 olan ve uzunluğunun genişliğine oranı (L/w) 0.8 olan dikdörtgensel bir kavite ele alınmıştır. Yaklaşmakta olan akımın tipi laminer olarak seçilmiştir. Kavite derinliği ve serbest akış hızına göre hesaplanan Reynolds sayısı 4000’dir. LES kullanılarak yapılan sayısal çalışmanın doğruluğu, 2 farklı Re sayısı için (4000 ve 13000) için Özsoy ve diğerleri (2005)’nin yapmış olduğu deneysel çalışma kullanılarak sağlanmıştır. Doğrulama yapıldıktan sonra 2B’lu bir kavite üzerindeki akımın 3B’lu yapısı ile, 3B’lu bir kavite üzerindeki akış yapısı sırasıyla incelenmiştir. Hem 2B’lu hem de 3B’lu kavite akışının içerdikleri akım ayrılması, kayma tabakası, vorteks üretimi, vorteks yayınma mekanizması ve vorteks-duvar etkileşimleri nedeniyle oldukça karmaşık bir yapıya sahip oldukları gözlemlenmiştir. 2B’lu ve 3B’lu çözümlerin kavite orta düzleminde yapılan karşılaştırmaları ortalama akış büyüklükleri cinsinden benzer sonuçlar verse de türbülanslı büyüklükler açısından farklılıklar içerdiği ve özellikle duvara yakın bölgelerde duvar etkisi nedeniyle oldukça farklı bir akış yapısı sergiledikleri gözlemlenmiştir. Anahtar Kelimeler: Kavite akışı, LES, Vorteks, Laminer akım.There are many engineering structures which require some form of cavities on their surfaces in order to perform the task they are designed for. However, these cavities can also be a source of various problems if structures with cavities are subjected to flow conditions. It is well known that the flow over a cavity is a complex phenomenon due to there usually being strong pressure, velocity and sometimes density fluctuations. These in turn can lead to increased drag forces, excessive noise and vibrations problems in structures. It is therefore essential to understand the cavity-flow physics and to be able to model, analyze and predict the outcome of such flows with acceptable accuracy if the adverse effects are to be controlled and minimized. Although many studies have been conducted about the cavity flows most of them have mainly focused on supersonic and compressible flows because of their relevance to aeronautical applications. However, much less attention has been given to very low Mach number cavity flows that are encountered in various grounds, industrial and environmental applications so far. There have been some studies dealing with incompressible cavity flow, but these were mainly focused on flow inside or around two-dimensional cavities and there have been very few investigations on flow inside a three-dimensional cavity. Therefore, there is a real need for better understanding of the three-dimensional effects in cavity flows at very low Reynolds numbers. The main purpose of this work is to study the three-dimensional effects on the characteristics of the cavity flow at very low Mach numbers and to improve the understanding of this type of flow. Therefore this study presents a numerical study into the problem of incompressible laminar flow passing a two and three-dimensional rectangular cavity with a view to understanding the complex nature of the flow and the three-dimensional structure of the cavity. The study is based on the solution of the unsteady, three-dimensional, incompressible Navier-Stokes equations by using finite volume method and Large Eddy Simulation approach. A cavity with length to depth ratio of 4 under an incoming laminar boundary layer is investigated for Re is equal to 4000 (based on cavity depth and free stream velocity). To validate the used computational procedure and simulations, the study of Özsoy et al. (2005) on two dimensional cavities under laminar incoming flow with L/h ratio of 4 is used. After this validation step, three-dimensional flow features of incompressible cavity flows with an incoming laminar boundary layer is investigated. The influence of the recirculating flow inside the cavity on the separated shear layer, the nature of the interactions between the large-scale eddies and the trailing edge corner and the dynamic behavior of the vortex structures occurring in a cavity are studied. The results obtained from computations show that the flow field inside the cavity has a three dimensional structure, even in two-dimensional cavity geometries including vortex shedding mechanism, stretching, wall vortex interaction like clipping or partially escape and so on. Time-averaged parameters in 3D analyses of 3D cavity reveal that the characteristics of the flow change significantly due to the side wall effects. As expected the wall effects are stronger at locations closer to the wall. Another important result from 3D analyses is that if the instantaneous flow field is considered, flow structures are quite different at different stations and highly time dependent. . When 3D flow characteristics are compared to those of 2D it is found that the mean flow features are almost the same for 3D and 2D only at the middle of the cavity. However, this is not the case when the results are compared outside this plane. The 2D and 3D results are significantly different at planes away from the symmetry plane. Although the mean flow quantities are about the same for 2D and 3D in the middle plane, turbulence behavior is quite different. It is believed that the validated computational procedure presented in this study for the analyses of flow over cavities can be used to determine the forces acting to the structures due cavities and this information can also be used for design purposes. Also, the validated computational procedure in this work can help future researchers to perform similar analyses for various kinds of cavities with high level of confidence and to optimize the geometry of the cavities for minimum drag, noise and vibration. Keywords: Cavity flow, vortex, LES, laminar flow.
Uygurcadan Türkçeye bilgisayarlı çeviri
Get PDFMachine translation is a sub-field of Natural Language Processing which belongs to Artificial Intelligence. Generally, it is based on computer technology that uses software to translate one natural language to another. In the 1950s, the Georgetown experiment involved fully-automatic translation of over sixty Russian sentences into English (Hutchins, 2004). The experiment was a great success and ushered in an era of substantial funding for machine-translation research. One of the main projects initiated by the US at that time was a machine translation system which converted Russian to English. This project continued from 1950 to 1960. In 1964, government sponsors of machine translation in the United States formed the Automatic Language Processing Advisory Committee (ALPAC) to examine the project's potential. In the famous 1966 report, ALPAC concluded that machine translation was slower, less accurate and twice as expensive as human translation, and that "there is no immediate or predictable prospect of useful machine translation" (Hutchins, 1995). The effects of this report brought about the virtual end to machine translation research in the US for over a decade after its publication. As computer technology developed, high capacity and high speed computers were produced. Thus, the main restrictions of studying natural language were removed and machine translation gained the attention of the computer science community once again. Despite technologic advances and the advent of new methods, a general purpose for full automatic machine translation systems still does not exist. To date, few machine translation systems have been developed, furthermore, they may only be applied to restricted texts and some post-editing works (usually necessary after initial translations). The main reasons for these are the morphological, syntactical and lexical differences between different languages. In conclusion, translated texts remain inferior to higher quality translations. Recently, some machine translation systems designed for related languages, such as: Czech to Slovak, Spanish to Catalan, and Turkmen language to Turkish have been implemented; studies on them have proven successful translations can be produced efficiently. In this study, our aim was to implement a machine translation system between Uyghur language and Turkish. Uyghur language is an agglutinative language such as other Turkic languages (i.e. Turkmen, Kazakh, Kyrgyz, Uzbek and Azeri etc.). All Turkic languages belong to the Ural-Altaic language family and are characteristically agglutinative languages which have productive inflectional and derivational morphology. Most research about natural language processing and machine translation of Turkic languages focus on Turkish language. Mainly due to the fact that there is active ongoing research on the subject in Turkey, and they continue to produce valuable results. To date, machine translation systems implemented between Turkic languages has been scant, such as: Turkish to Azeri, Turkish to Crimean Tatar, Turkmen language to Turkish etc. Unfortunately, little computational research about Uygur languages exists. Turkic languages tend to have similar morphological structure and share some common word roots. The main shared properties include similar word order and syntactic structure. However, distinctions exist which prevent mutual intelligibility between these languages. In order to implement this translation system, we utilized a frame-work which is favored for translation between closely related agglutinative languages. Thus, we implemented a morphological analyzer for Uyghur language with XEROX's Finite State Transducers (FST) tools. In this morphological analyzer we considered general cases for Uyghur languages and tagged Uyghur words with the same tags that were used for tagging other Turkic languages words. Thus, it will be easy to integrate this system to other Turkic languages. In order to improve the system's performance, we implemented a rule based morphological disambiguator, additionally, a disambiguator for word senses. We have evaluated our system's performance using BLEU scores for 240 differently structured sentences. As a result, a system has been determined which may successfully translate intermediate level Uyghur language into Turkish. Keywords: Machine translation, Turkic languages, Uyghur language, Turkish.Bilgisayarlı Çeviri (BÇ) yapay zeka çalışmalarının bir alt dalı olan Doğal Dil İşlemenin (DDİ) alt konusudur. Diller arası çeviride bilgisayarların kullanılması fikri 1950’lerin ilk yıllarında ortaya çıkmıştır. O tarihten günümüze kadar pek çok dil üzerinde çalışılmış ve çeşitli yöntemler geliştirilmiştir. Ancak teknolojideki ve yöntemlerdeki gelişmelere karşın, genel amaçlı, yüksek başarıma sahip çeviri sistemleri henüz geliştirilememiştir. Bunun temel nedeni, diller arasındaki büyük yapısal ve anlatım farklılıklarıdır. Yapısal yönden benzer olan diller arasına bilgisayarlı çevirinin daha kolay olduğu bilinmektedir. Son yıllarda Çekçe-Slovakça, Çekçe-Lehçe, İspanyolca-Katalanca, Türkmence-Türkçe gibi yakın diller arasında yüksek başarımlı çeviri yapabilen sistemler geliştirilmiştir. Akraba veya yakın diler arasında çeviri amaçlı geliştirilen sistemler, farklılıkların büyük olduğu, Türkçe-İngilizce gibi diller arasında bilgisayarlı çeviri için gerek duyulan karmaşık yöntemlere göre, daha basit ve kolay gerçeklenebilir yöntemler kullanmaktadırlar. Bu çalışma kapsamında, aynı dil ailesi içinde sınıflandırılan ve birçok yönden benzerlikler gösteren Uygurcadan Türkçeye bilgisayarlı çeviri sistemi geliştirilmiştir. Aslında bu diller ne kadar benzer özellikler gösterse de, çözülmesi gereken farklılıklar azımsanmayacak kadar çoktur. Genel olarak Uygur Türkçesi ile Türkiye Türkçesinin söz dizimi aynıdır. Bundan dolayı çeviri sistemi geliştirirken, sözcüklerin dizilimi değişmemektedir. Ancak sözcüklere eklenen ekler çok farklılaşabilmektedir. Uygurca ve Türkçe bitişken diller olduğundan, ekler çok önemlidir. Ekler sözcüklerin hatta tümcenin anlamını değiştirmektedir. Bu çalışmada, akraba ve bitişen diller arasında bilgisayarlı çeviri için geliştirilen karma model üzerine, belirsizlik giderme yönteminin eklenmesi ile Uygurcadan Türkçeye bilgisayarlı çeviri sistemi geliştirilmiştir. Anahtar Kelimeler: Bilgisayarlı çeviri, Türk Dilleri, Uygurca, Türkçe
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!