Teknik Bilimler Meslek Yüksekokulu Koleksiyonu

Permanent URI for this collectionhttps://hdl.handle.net/20.500.13091/1629

Browse

Now showing 1 - 9 of 9

Citation - WoS: 12
Citation - Scopus: 17
Acoustic-Driven Airflow Flame Extinguishing System Design and Analysis of Capabilities of Low Frequency in Different Fuels
(Springer, 2022) Taşpınar, Yavuz Selim; Köklü, Murat; Altın, Mustafa
Traditional fire extinguishing methods can harm people and nature. For this purpose, in this study, no harmful acoustic-driven airflow fire extinguishing system was developed and experiments were carried out to extinguish gasoline, kerosene, thinner (synthetic thinner) liquid fuels and liquid petroleum gas (LPG) flames. 17,442 extinguishing experiments were conducted in 5 different flame sizes, 54 different frequencies and 10 cm to 190 cm distance range. The data obtained were analyzed using the polynomial regression method. For liquid fuels, the frequencies of 10 Hz to 50 Hz at a distance of 10 cm to 100 cm, 10 Hz to 32 Hz at a distance of 100 cm to 150 cm, and 10 Hz to 28 Hz at a distance of 150 cm to 180 cm are effective extinguishing ranges. LPG fuel, 10 Hz to 45 Hz at a distance of 10 cm to 140 cm, frequencies of 15 Hz to 30 Hz at a distance of 140 cm to 180 cm are effective extinguishing ranges. In addition, caused by the compression of the woofers membrane inside the collimator and the injected airflow at frequency of 30 Hz reduced the 12 cm diameter metal plate from 86.2 degrees C to 18.8 degrees C in 5 min, and the metal plate left to cool down from 80 degrees C to 21.7 degrees C in 10 min at a distance of 100 cm. The average Mean Square Error value obtained as a result of polynomial regression analysis is 0.9544, and the Root Mean Square Error value is 1.2020.
Cam Lif Takviyeli Betonun Yangın Dayanımlarının Çeşitli Parametreler Açısından İrdelenmesi
(2019) Ali, Ali Murthada; Arslan, Musa Hakan; Altın, Mustafa
Cam lifi taze beton karışımına eklenerek betonun mekanik özellikleri iyileştirilmektedir. Literatürde cam lif takviyeli betonun (CTB) mekanik özelliklerinin tespitine yönelik çok sayıda araştırma yapılmıştır. Beton içine cam lifi eklenerek özellikle betonun eğilme ve çekme dayanımında hatırı sayılır artışlar meydana gelmekte, bu şekilde üretilmiş olan betonlar daha ince kesitli oldukları için de uygulamada tercih edilmektedirler. Betonun içine eklenen cam lifinin betonun yangın performansına olumsuz bir etki yapıp yapmadığına yönelik kapsamlı bir araştırma olmadığı için bu çalışmada dört farklı (30,60, 90 ve 120 dk) yangın sürelerine tabi tutulmuş cam lif takviyesiz beton ile CTB elemanlarının yangın sonrasındaki mekanik özelliklerindeki değişiklikler incelenmiştir. Deneysel çalışma kapsamında Ø15/30 cm silindir ve 4x4x16 cm prizmadan oluşan toplam 300 adet numune üretilmiştir. Numunelerde kullanılan 0,5,10,15 ve 20 kg/m3 oranlarındaki cam elyaf takviyeli betonların değişik sürelerde yangın sonrasındaki basınç ve üç farklı çekme (yarma, tek noktadan eğilme ve iki noktadan eğilme) dayanımı üzerindeki etkisi farklı sürelerde yangına maruz bırakılan CTB elemanları üzerinden test edilmiştir. Yangın geçirmemiş numunelerde cam lifi basınç dayanımına bir katkıda bulunmazken çekme dayanımında önemli oranlarda katkısı olmuştur. 30 dakikalık yangın sonrasında CTB numunelerin tek ve çift noktadan eğilme deneylerinden elde edilen çekme dayanımlarında yangın öncesine göre bir artış görülmüştür. Daha uzun süreli yangınlarda (60, 90 ve 120 dk) ise bu oranda bir artış meydana gelmemiştir.
Citation - WoS: 20
Citation - Scopus: 29
Classification of Flame Extinction Based on Acoustic Oscillations Using Artificial Intelligence Methods
(ELSEVIER, 2021) Taşpınar, Yavuz Selim; Köklü, Murat; Altın, Mustafa
Fire, one of the most serious disasters threatening human life, is a chemical event that can destroy forests, buildings, and machinery within minutes. For this reason, there have been numerous methods developed to extinguish the fire. Within the scope of this study, a sound wave flame extinction system was developed in order to extinguish the flames at an early stage of the fire. The data used in the study were obtained as a result of experiments conducted with the developed system. The created dataset consists of data obtained from 17,442 experiments. It is aimed to classify the fuel type, flame size, decibel, frequency, airflow and distance features, and the extinction-non-extinction status of the flame through rule-based machine learning methods. In the study, rule-based machine learning methods, ANFIS (Adaptive-Network Based Fuzzy Inference Systems), CN2 Rule and DT (Decision Tree) were used. The methods of Box Plot, Scatter Plot and Correlation Analysis were utilized for statistical analysis of the data. As a result of the classifications, respectively, 94.5%, 99.91%, and 97.28% success were achieved with the ANFIS, CN2 Rule, and DT methods. As a result of the evaluations made by using Box Plot, Scatter Plot and Correlation Analysis.
Citation - Scopus: 18
Fire Detection in Images Using Framework Based on Image Processing, Motion Detection and Convolutional Neural Network
(Ismail Saritas, 2021) Taşpınar, Yavuz Selim; Köklü, Murat; Altın, Mustafa
Fire detection in images has been frequently used recently to detect fire at an early stage. These methods play an important role in reducing the loss of life and property. Fire is not only chemically complex, but also physically very complex. The shape and color of the flame varies according to the type of fuel in the fire. This has made fire detection a very challenging problem. Advanced image processing algorithms are also needed to accurately detect fire. To solve this problem, a three-stage fire framework was created in this study. In the first stage, the flame region was extracted from the images containing the fire region with the basic image processing algorithms. At this stage, reduce brightness, HSL, YCbCr, median and herbaceous filters are applied successively to the image. Since the flame image has a polygonal structure by nature, the number of edges of the flame region has been found. In the second stage, the mobility feature of the flame was utilized. For this purpose, the mobility of the flame was determined by comparing the video frames containing the fire image. The CNN method was used to detect the fire in the images. The CNN model was trained with the transfer learning method using the Inception V3, SequeezeNet, VGG16 and VGG19 trained models. As a result of the tests of the models, 98.8%, 97.0%, 97.3% and 96.8% classification success were obtained, respectively. With the proposed fire detection framework, it is thought that the damage caused by the fire can be reduced by early detection of the fire and timely intervention. © 2021, Ismail Saritas. All rights reserved.
Citation - Scopus: 10
Identification of the English Accent Spoken in Different Countries by the K-Nearest Neighbor Method
(2020) Taşpınar, Yavuz Selim; Köklü, Murat; Altın, Mustafa
Sound is the pressure wave created by an object vibrating with a certain frequency. 3 organs are needed for the formation of voice in humans. These are lungs, vocal cords and mouth. Due to the structure of these organs and the similarity of the person with their current language, they can speak another language with different accent. A language can be spoken in different parts of the same country and in different countries. The second most widely used language in the world is English, has numerous accents around the world. In this study, it is aimed to determine which country the English accent spoken in different regions belongs to. In the dataset used, there are 330 sound samples including English accents spoken in Spain, France, Germany, Italy, England and America. Classification has been made with 12 features obtained by Mel Frequency Cepstrum Coefficients feature extraction method. k-Nearest Neighbor (kNN) were used in the classification and 87.2% success was achieved.
Citation - WoS: 20
Citation - Scopus: 23
Investigation of Usability of Boron Doped Sheep Wool as Insulation Material and Comparison With Existing Insulation Materials
(Elsevier Sci Ltd, 2022) Altın, Mustafa; Yıldırım, Giray Şamil
The limited energy resources in the world require maximum attention in energy saving and energy use. It has always been important to conserve energy by making thermal insulation in buildings which are one of the places where energy is used the most, and subsequently reducing environmental pollution as a result. The importance of existing thermal insulation materials in maintaining heat is already known. However, the damage they cause to the environment in their production is great, therefore insomuch that the manufacture and use of some insulation materials are prohibited. As a result, it is known that studies have been carried out to produce alternative thermal insulation materials. In this study, as a thermal insulation material, the usability of sheep wool which is an environmentally friendly, economical, and natural fiber, has been researched. A new insulation board was produced by adding boron to sheep wool, and this new insulation board was compared with rock wool and expanded polystyrene (EPS), which are widely used today. To carry out the experimental studies, an experiment building with full size and 3 rooms had been built. First, the building was completely covered with rock wool and the effect of the facades on the temperature and moisture in the environment was observed. Afterwards, as one room was covered with EPS and the other room was covered by the boron doped sheep wool insulation material, the insulation materials were compared wtih one another. In the experimental studies that was carried out, while rock wool gave the best results in ambient and surface temperature tests, EPS provided the best insulation in the moisture test. In the sound test, while rock wool and EPS absorbed the sound at very close decibel levels, the boron-doped sheep wool insulation material provided a higher level of sound insulation. In the experimental studies that was carried out, it has been determined that the boron-doped sheep wool insulation material has serious competition with both rock wool and EPS even with the first production of 10 mm.
Citation - Scopus: 18
Object Recognition With Hybrid Deep Learning Methods and Testing on Embedded Systems
(Ismail Saritas, 2020) Taşpınar, Yavuz Selim; Selek, Murat
Object recognition applications can be made with deep neural networks. However, this process may require intensive processing load. For this purpose, hybrid object recognition algorithms that can be created for the recognition of an object in the image and the comparison of the working time of these algorithms on various embedded systems are emphasized. While Haar Cascade, Local Binary Pattern (LBP) and Histogram Oriented Gradients (HOG) algorithms are used for object detection, Convolutional Neural Network (CNN) and Deep Neural Network (DNN) algorithms are used for classification. As a result, six hybrid structures such as Haar Cascade+CNN, LBP+CNN, HOG+CNN and Haar Cascade+DNN, LBP+DNN, HOG+DNN are developed. In this study, these 6 hybrid algorithms were analyzed in terms of success percentage and time, then compared with each other. Microsoft COCO dataset was used to train and test all these hybrid algorithms. Object recognition success of CNN was 76.33%. Object recognition success of Haar Cascade+CNN, one of the hybrid methods we recommend, with a success rate of 78.6% is higher than CNN and other hybrid methods. LBP+CNN method recognized objects in 0.487 seconds which is faster than any other hybrid methods. In our study, Nvidia Jetson TX2, Asus TinkerBoard, Raspbbery Pi 3 B+ were used as embedded systems. As a result of these tests, Haar Cascade+CNN method on Nvidia Jetson TX2 was detected in 0.1303 seconds, LBP+DNN and Haar Cascade+DNN methods on Asus Tinker Board were detected in 0.2459 seconds, and HOG+DNN method on Raspberry Pi 3 B+ was detected in 0.7153 seconds.. © 2020, Ismail Saritas. All rights reserved.
Predicting Student Dropout Using Machine Learning Algorithms
(2024) Sulak, Süleyman Alpaslan; Köklü, Niğmet
This article comprehensively examines the use of machine learning algorithms to predict and reduce student dropout rates. These methods, developed to monitor and support student achievement in education, also aimedto enhance success rates in education and ensure more effective student engagement in the learning process. Bigdata analysis and machine learning models provide important contributions to the development of strategic solutions to the problem of school dropout by predicting student movements and trends.This study uses a dataset consisting of 4424 student data and has 37 features. The dataset is divided into three classes: "Dropout", "Enrolled" and "Graduate" according to the students' school dropout status. Decision Tree (DT), Random Forest (RF) and Artificial Neural Network (ANN) competitions, which are frequently used in such training studies in the literature, are aimed at this dataset. According to the obtained operations, DT showed moderate performance with an accuracy rate of 70.1%. The RF algorithm showed higher success with an accuracy rate of 75.5%. The highest success was achieved by the ANNalgorithm with an accuracy rate of 77.3%. ANN's flexible structure has produced superior results compared to other algorithms for this dataset, its ability provide successful classification in complex datasets.The article ultimately demonstrates how machine learning-based prediction models can have a significant impact on student achievement and offer a powerful tool for reducing school dropouts.
World Geography With Augmented Reality
(2021) Köklü, Niğmet; Sulak Süleyman Alpaslan
Mobile technologies enable individuals to organize and manage their daily lives through the mobile applications that are used during the day. Augmented reality is one of those applications that can be used on mobile devices. Expectations for the use of new technologies in educational environments increase in parallel with the technological advancements. Thanks to advances in computer, internet and mobile technologies, numerous new applications have emerged that facilitate communication and interaction in educational environments. In particular, the applications of the augmented reality technologies emerging with the advancing technology in the field of education are becoming widespread and easy-to-use. Augmented reality is a technology that enriches the learning process, created by real-time integration of the virtual elements and the real world elements. In geography teaching, different knowledge and skills are expected to be acquired by the learner, therefore augmented reality applications appealing to the different senses of learners can be utilized. AR technology can be used in various fields, especially in education. In addition to increasing productivity, mobile augmented reality applications would be used for multiple purposes in the field of education with the interactive facilities they offer to user such as 2D and 3D visual support and video playback.

Browse

Browsing Teknik Bilimler Meslek Yüksekokulu Koleksiyonu by Department "Meslek Yüksekokulları, Teknik Bilimler Meslek Yüksekokulu, İnşaat Bölümü"