Mekatronik Mühendisliği Bölümü Makale Koleksiyonu

Bu koleksiyon için kalıcı URI

https://hdl.handle.net/20.500.12418/304

Listeleniyor 1 - 13 / 13

Experimental and Numerical Investigation of the Control of the Flow Structure on Surface Modified Airfoils
(2023) Öztürk,Adnan; Çoban, Mehmet; Koca, Ferhat
In this study, experimental and numerical flow analysis was performed on three different blade profiles with a chord length of 165 mm using passive flow control method. The first of the airfoil is the standard NACA 0018 profile. The second airfoil type has a NACA 0018 profile with a gap in the suction surface. The last airfoil is the NACA 0018 profile which is 66% of the trailing edge cut from the chord length. All airfoil profiles were analyzed at the Reynolds number, Re=2x104, and angles of attack α=0o, 5o, 10o, 12o and 14o in both experiment and numerical studies. The experiments were carried out using the Particle Image Velocimetry (PIV) method in a closed-loop open water channel, and the time-averaged velocity vectors, streamlines, and vorticity contours of the flow field were obtained. Subsequently, numerical analyses were performed using the ANSYS Fluent package program, one of the Computational Fluid Dynamics (CFD) programs used frequently in the literature. The streamlines and pressure contours of the airfoil profiles have been compared visually at the same Re and different angles of attack. In addition, according to the angle of attack of the airfoil profiles, lift coefficient CL, drag coefficient CD, and the ratio of lift coefficient to drag coefficient CL/CD graphs were presented. It has been shown that the gap on the airfoil at high attack angles caused changes in lift (up to 0.7) and drag (up to 0.15). These features can allow these models to be used for different purposes in the aerodynamics field.
Experimental investigation of the effects of boron oil additive in internal combustion engine
(Elsevier, 2023) Karataş, Ömer; Yüksel, Tahsin
Using additives to engine oils increases engine performance, extending engine maintenance life, boosting the competitiveness of mineral-based oils technically and economically. In this study, the effects of liquid boron oil additive added to mineral-based 10 W-40 engine lubricating oil used in a single-cylinder diesel engine were investigated. 15 % of the engine oil volume was added to the engine as an additive. Experimental studies were carried out at three different engine speeds (i.e., 1500, 1750, and 2000 rpm). The engine, on which no changes were made, was first run with mineral lubricating oil, and then the experiments were repeated with mineral lubricating oil with 15 % boron additive. In the studies, engine torques, specific fuel consumption, and exhaust emissions (HC, CO, CO2, and NOx) were evaluated. In addition, the effects of operating the engine for 100 h without adding any additives and with oil additives containing 15 % boron were also investigated. It was found that engine torque and EGT increased while BSFC, CO2, HC, and NOx decreased in boron-added mineral lubricating oil for all revolutions. In addition, there was a general decrease in CO emissions. In-cylinder SEM images and oil analyses, which were operated with a 15 % boron additive to the engine oil, showed that the boron additive positively affected the engine.
Automatic Detection and Mapping of Dolines Using U-Net Model from Orthophoto Images
(2023) Polat, Ali; Keskin, İnan; Polat, Özlem
A doline is a natural closed depression formed as a result of karstification, and it is the most common landform in karst areas. These depressions damage many living areas and various engineering structures, and this type of collapse event has created natural hazards in terms of human safety, agricultural activities, and the economy. Therefore, it is important to detect dolines and reveal their properties. In this study, a solution that automatically detects dolines is proposed. The proposed model was employed in a region where many dolines are found in the northwestern part of Sivas City, Turkey. A U-Net model with transfer learning techniques was applied for this task. DenseNet121 gave the best results for the segmentation of the dolines via ResNet34, and EfficientNetB3 and DenseNet121 were used with the U-Net model. The Intersection over Union (IoU) and F-score were used as model evaluation metrics. The IoU and F-score of the DenseNet121 model were calculated as 0.78 and 0.87 for the test data, respectively. Dolines were successfully predicted for the selected test area. The results were converted into a georeferenced vector file. The doline inventory maps can be easily and quickly created using this method. The results can be used in geomorphology, susceptibility, and site selection studies. In addition, this method can be used to segment other landforms in earth science studies.
TIG-weldability of AISI 430 and DUROSTAT 500 grade
(De Gruyter, 2023) AYDIN Sinan
In this study, 10mm thick DUROSTAT 500 and AISI 430 grades were joined by double sided keyhole tungsten inert gas (K-TIG) welding method without using filler material. The characterization of the microstructure of the weld zone was investigated by optical analysis methods and the mechanical properties of the welded parts were examined by mechanical tests. The fracture surface structure of the parts that were broken as a result of the tests were examined. No deterioration was observed in the welded samples. It was determined that the weld penetration increased as a result of the increase in the amount of heat entering the weld zone with the increasing welding current.
AISI 1040 ve AA6013 malzeme çiftinin mekanik kilitleme yönteminde (MLM) farklı bağlantı açıları kullanılarak birleştirilmesi
(Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, 2022) Mercan, Serdar; Özkavak Varol, Hatice
Farklı malzeme türlerinin birleştirilerek kullanılmasının zorunlu olduğu birçok endüstriyel uygulama bulunmaktadır. Ancak kaynak, döküm, yapıştırma gibi diğer birleştirme yöntemleri ile bu her zaman mümkün olmamakta ya da çeşitli problemler ortaya çıkarmaktadır. Bu problemlerin ortadan kaldırılması amacıyla klasik birleştirme yöntemlerine alternatif olarak yeni ve çevreci bir yöntem olan mekanik kilitleme yöntemi (MLM) kullanılmaktadır. MLM yöntemi özellikle farklı kimyasal ve fiziksel özelliklere sahip malzeme çiftlerinin birleştirilmesinde kullanılabilir. Bununla birlikte yüksek ısı girdisinin neden olduğu mikro yapı hatalarının azaltılması istenen durumlarda tercih edilebilir. Bu çalışmada; MLM için bağlantı kalitesini doğrudan etkileyen bağlantı açısının mikro yapı ve mekanik özelliklerle olan etkisi araştırılmıştır. Bu amaçla; AISI 1040 çelik ve AA6013 alüminyum alaşımı MLM yöntemi ile farklı bağlantı açıları (7°,10° ve 13°) kullanılarak birleştirilmiştir. Birleştirme işlemi sonrasında mikro yapı incelemesi, mikro sertlik ölçümleri ile çekme testleri yapılarak bağlantı performansı belirlenmiştir. Çalışma sonunda 13° bağlantı açısı ile hazırlanan numunelerde 241,38 MPa değer ile en yüksek mekanik özellikler elde edilmiştir. Bağlantı açısının daha düşük olduğu diğer numunelerde ise bağlantıyı gerçekleştiren flanş oluşumunun düzensiz dağılım gösterdiği, malzeme akışının yetersiz olduğu bunun sonucunda mekanik özelliklerin düştüğü tespit edilmiştir.
Reducing the Roughness and Sound Intensity by Optimization of Cutting Parameters in Processing of AISI 2714 Steel Material on CNC Milling Machine
(2022) Mercan, Serdar
Within the scope of this study, optimization of cutting parameters (feed rate, cutting speed and depth of cut) was aimed in order to reduce the noise level that occurs during the processing of AISI 2714 steel on CNC milling machine without compromising the surface roughness. Experimental design was examined in three variables, three levels and two target functions. In order to investigate the contribution of these parameters to the target function, the experiments were carried out in accordance with the experiment plan determined by using the "Central Composite Design (CCD)" of the "Response Surface Method (RSM)". Mathematical models have been developed in order to predict sound intensity and surface roughness by applying regression analysis to the experimental results. As a result, it has been observed that the most effective parameter in reducing the surface roughness is the feed rate, followed by the depth of cut. While the depth of cut was the most effective parameter in reducing the sound intensity, it was determined that the next effective parameter was the feed rate.
Brain tumor classification by using a novel convolutional neural network structure
(2022) Polat, Özlem; Dokur, Zümray; Ölmez, Tamer
Brain tumors located in the skull are among the health problems that cause serious consequences. Rapid and accurate detection of brain tumor types will ensure that the patient receives appropriate treatment in the early period, thus increasing the patient's chance of recovery and survival. In the literature, classification accuracies over 98% have been acquired automatically by using deep neural networks (DNN) for the brain tumor images such as glioma, meningioma, and pituitary. It is observed that researchers generally focused on achieving higher classification accuracy and therefore, they have used pre-processing stages, augmentation processes, huge or hybrid DNN structures. These approaches have brought some disadvantages in terms of practical use of the developed methods: (i)The parameters of the pre-processes should be carefully determined, otherwise the classification accuracy will decrease. (ii) In order to increase the classification performance, it is important to determine the coarse structure of the DNN correctly. If the DNN has many hyper-parameters, the coarse structure will be determined in a long time. (iii) It is difficult to implement complex DNN structures or training algorithms in terms of practical use, because these methods need huge memory and high CPU computation. In this study, we have proposed a novel DNN model to increase the classification accuracy, and to decrease the number of weights in the structure, and to use less number of hyper-parameters. We named this model, which uses a divergence-based feature extractor, as DivFE-v1 for short. 99.18% classification accuracy for the Figshare dataset is obtained by using the small-sized DNN structure without any pre-processing stage or augmentation process.
Determination of COPD severity from chest CT images using deep transfer learning network
(2022) Polat, Özlem; Şalk, İsmail; Doğan, Ömer Tamer
The purpose of this study is to present a solution to the problem of detecting the severity of Chronic Obstructive Pulmonary Disease (COPD) from chest CT images using deep transfer learning network. The study has a novelty in terms of classifying the severity of COPD with machine learning methods for the first time in the literature. Transfer learning has been preferred because of its proven performance in image analysis and classification. In this study, a dataset containing a total of 1815 CT images from 121 patients with moderate, severe and very severe COPD was used. Lung parenchyma was first segmented from CT images using HSV color space thresholding. Then Inception-V3 model was trained and tested on the segmented image dataset for COPD severity classification. The tests were repeated 10 times. The proposed model was able to detect the severity level of COPD with an average accuracy of 96.79% and a maximum of 97.98%. The classification result proved that the severity of COPD can be classified with very high performance. Thus, the applied transfer learning is promising in medical sciences and can assist to radiologists in making quick and accurate decisions.
Classification of brain tumors from MR images using deep transfer learning
(2021) Polat, Özlem; Güngen, Cahfer
Classification of brain tumors is of great importance in medical applications that benefit from computer-aided diagnosis. Misdiagnosis of brain tumor type will both prevent the patient from responding effectively to the applied treatment and decrease the patient’s chances of survival. In this study, we propose a solution for classifying brain tumors in MR images using transfer learning networks. The most common brain tumors are detected with VGG16, VGG19, ResNet50 and DenseNet21 networks using transfer learning. Deep transfer learning networks are trained and tested using four different optimization algorithms (Adadelta, ADAM, RMSprop and SGD) on the accessible Figshare dataset containing 3064 T1-weighted MR images from 233 patients with three common brain tumor types: glioma (1426 images), meningioma (708 images) and pituitary (930 images). The area under the curve (AUC) and accuracy metrics were used as performance measures. The proposed transfer learning methods have a level of success that can be compared with studies in the literature; the highest classification performance is 99.02% with ResNet50 using Adadelta. The classification result proved that the most common brain tumors can be classified with very high performance. Thus, the transfer learning model is promising in medicine and can help doctors make quick and accurate decisions.
Automatic classification of volcanic rocks from thin section images using transfer learning networks
(2021) Polat, Özlem; Polat, Ali; Ekici, Taner
In this study, efficient deep transfer learning models are proposed to classify six types of volcanic rocks, and this paper has a novelty in classifying volcanic rock types for the first time using thin section images. Convolutional neural network-based DenseNet121 and ResNet50 networks, which are transfer learning methods, are used to extract the features from thin section images of rocks, and the classification process is carried out with a single-layer fully connected neural network. The proposed models are trained and tested on 1200 thin section images using four different optimizers (Adadelta, ADAM, RMSprop, SGD). AUC, accuracy, precision, recall and f1-score are used as performance metrics. Proposed models are run 10 times for each optimizer. DenseNet121 classifies volcanic rock types using RMSprop with an average accuracy of 99.50% and a maximum of 100.00%, and ResNet50 classifies using ADAM with an average accuracy of 98.80% and a maximum of 99.72%. Thus, the applied deep transfer learning is promising in geosciences and can be used to identify rock types quickly and accurately.
Classification of plutonic rock types using thin section images with deep transfer learning
(2021) Polat, Özlem; Polat, Ali; Ekici, Taner
Classification of rocks is one of the basic parts of geological research and is a difficult task due to the heterogeneous properties of rocks. This process is time consuming and requires sufficiently knowledgeable and experienced specialists in the field of petrography. This paper has a novelty in classifying plutonic rock types for the first time using thin section images; and proposes an approach that uses the deep learning method for automatic classification of 12 types of plutonic rocks. Convolutional neural network based DenseNet121, which is one of the deep learning architectures, is used to extract the features from thin section images of rocks; and the classification process is carried out with a single layer fully connected neural network. The deep learning model is trained and tested on 2400 images. AUC, accuracy, precision, recall and f1-score are used as performance measure. The proposed approach classifies plutonic rock images on the test set with an average accuracy of 97.52% and a maximum of 98.19%. Thus, the applied deep transfer learning is promising in geosciences and can be used to identify rock types quickly and accurately.
Determination of Pneumonia in X-ray Chest Images by Using Convolutional Neural Network
(2021) Polat, Özlem; Dokur, Zümray; Ölmez, Taner
Pneumonia is one of the major diseases that cause a lot of deaths all over the world. Determining pneumonia from chest X-ray (CXR) images is an extremely difficult and important image processing problem. The discrimination of whether pneumonia is of bacterium or virus origin has also become more important during the pandemic. Automatic determination of the presence and origin of pneumonia is crucial for speeding up the treatment process and increasing the patient’s survival rate. In this study, a convolutional neural network (CNN) framework is proposed for detection of pneumonia from CXR images. Two different binary CNNs and a triple CNN are used for determining: (i) normal or pneumonia, (ii) pneumonia of bacterium or virus origin, and (iii) normal or bacterial pneumonia or viral pneumonia. In this approach, CNNs are trained with Walsh functions to extract the features from CXR images, and minimum distance classifier instead of a fully connected neural network is employed for classification purpose. Training with Walsh functions maintains the within-class scattering to be low, and between-class scattering to be high. Preferring the minimum distance classifier reduces the number of nodes used and also allows the network to be controlled with fewer hyperparameters. These approaches bring some advantages to the system designed for the classification process: (i) easy determination of hyperparameters, (ii) achieving higher classification performance, and (iii) use of fewer neurons. The proposed smallsize CNN model was applied to CXR images from 1- to 5-year-old children provided by the Guangzhou Women’s and Children’s Medical Center (GWCMC). Three experiments have been conducted to improve the classification performance: (i) the effect of different sizes of input images on the performance of the network was investigated, (ii) training set was augmented by randomly flipping left to right or down to up, by adding Gaussian noise to the images, by creating negative images randomly, and by changing image brightness randomly (iii) instead of RGB CXR images, gray component of the original image and four 2D wavelet images were given as input to the network. In these experiments, no major changes were observed in the classification results obtained by using the proposed CNNs. The proposed method has achieved 100% accuracy for normal or pneumonia, 92% for pneumonia of bacterium or virus origin, and 90% for normal or bacterial pneumonia or viral pneumonia. It is observed that higher classification performances were obtained with approximately five times less parameters compared to the networks that gave the best results in the literature. Thus, the applied CNN model is promising in medicine and can help experts make quick and accurate decisions.
A Joining Dissimilar Material Pairs by Mechanical Locking Method (MLM)
(International Journal of Precision Engineering and Manufacturing, 2021) Mercan, Serdar
Several methods that have been used in joining material pairs with diferent properties share many disadvantages. To overcome these disadvantages, the current study introduces mechanical locking method (MLM), which is a new and environmentally friendly method. This method can prevent many problems, especially those related to chemical incompatibility in practical applications. The method helps materials with physical and chemical incompatibilities join successfully. The only limitation in the joining process is that one of the materials must melt. When this criterion is met, ceramics can be joint with metals, ferrous-based materials can be joint with non-ferrous metals and other types of materials. This study used two diferent metal alloys, namely plain carbon steel and brass (CuZn30) for MLM as mold metal and reshaping metal, respectively. MLM process resulted in a successful joining for steel-brass alloys. Joint material pairs were examined using microstructural methods and mechanical properties. The study used the tensile test to evaluate the mechanical properties of joint metals, and used Vickers indentation for hardness measurement. The researcher conducted Scanning Electron Microscope (SEM) investigations for possible interface reactions and re-formed grain structures for both alloys.

Güncel Gönderiler