Person: KIRCI, PINAR
Loading...
Email Address
Birth Date
Research Projects
Organizational Units
Job Title
Last Name
KIRCI
First Name
PINAR
Name
7 results
Search Results
Now showing 1 - 7 of 7
Publication Transfer learning enabled bearing fault detection methods based on image representations of single-dimensional signals(Springer, 2024-02-21) Deveci, Bilgin Umut; Çeltikoğlu, Mert; Albayrak, Özlem; Ünal, Perin; Kırcı, Pınar; Çeltikoğlu, Mert; KIRCI, PINAR; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/Bilgisayar Mühendisliği Bölümü.; HXH-8622-2023; CZK-0182-2022Bearings are vital components in rotating machinery. Undetected bearing faults may result not only in financial loss, but also in the loss of lives. Hence, there exists an abundance of studies working on the early detection of bearing faults. The rising use of deep learning in recent years increased the number of imaging types/neural network architectures used for bearing fault classification, making it challenging to choose the most suitable 2-D imaging method and neural network. This study aims to address this challenge, by sharing the results of the training of eighteen imaging methods with four different networks using the same vibration data and training metrics. To further strengthen the results, the validation dataset size was taken as five times the training dataset size. The best results obtained is 99.89% accuracy by using Scattergram Filter Bank 1 as the image input, and ResNet-50 as the network for training. Prior to our work, Scattergram images have never been used for bearing fault classification. Ten out of 72 methods used in this work resulted in accuracies higher than 99.5%.Publication Determination of the fractal dimension of the active fault data along the east anatolian fault zone(Inst Geology & Geography, 2021-01-01) Bayrak, Ebru Aydındağ; Kırcı, Pınar; KIRCI, PINAR; Bursa Uludağ Üniversitesi/Bilgisayar Mühendisliği Bölümü; CZK-0182-2022The current study has analyzed active fault data along the East Anatolian Fault Zone (EAFZ) apply ing both manual (classic) and modem versions of the box counting method. The EAFZ active fault datasct used for analysis was taken from the Geoscience Map Viewer and the Drawing Editor from the website of the General Directorate of Mineral Research and Exploration. The study covered an area stretching from Karhova in the north to Kinkhan in the south. The fractal analysis of the earthquake surface rupture and the Holocene fault data was performed. Fractal dimensions of the EAFZ active-fault data were calculated for 15 boxes and compared with correlation coefficient values. The calculated fractal dimension values were found to vary with the density of the active-fault data falling into the boxes. The maximum fractal dimension value D-1 was determined for Karliova and its surroundings, which can be associated with the fault density due to the branching geometry.Publication A communication, management and tracking mobile application for enhancing earthquake preparedness and situational awareness in the event of an earthquake(MDPI, 2023-01-01) Kırcı, Pınar; Arslan, Davut; Dinçer, Süreyya Fatih; KIRCI, PINAR; Arslan, Davut; Dinçer, Süreyya Fatih; Bursa Uludağ Üniversitesi/Bilgisayar Mühendisliği Bölümü; CZK-0182-2022; HME-3490-2023; HMI-2917-2023The presented DepApp is an application that provides information about the intensity, time and whereabouts of a recent earthquake. In addition, the presented application is a mobile application that contains emergency phone numbers and can provide general information about an earthquake which people can benefit from before and after the earthquake. With the presented application, people will be able to easily access information about earthquakes. Moreover, safe areas in which to shelter after an earthquake are offered to the user. Information about an earthquake kit is given to user in the application. The earthquake kit should be prepared before the earthquake. The application also includes emergency sounds that can be used in the event of an earthquake. A survey system is designed to determine the intensity of the earthquake as it is felt by the users. After the earthquake, the application presents the intensity, time, and location of the earthquake to the user. Thus, with this application, users will be more knowledgeable and prepared for an earthquake.Publication Analyzing blood donation probabilities and number of possible donors(Ieee, 2020-01-01) Kırcı, Pınar; KIRCI, PINAR; Aktaş, Şeyma; Sevinç, Burcu; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/Bilgisayar Mühendisliği Bölümü.In the paper, many critical data of donors were used. A donor's blood donation frequency and the last donation time were included in the utilized data. The utilization of these data types were very important for providing a solution to determine blood donation probabilities. By using many machine learning approaches on blood transfusion data, it is tried to be estimated, if a possible donor will provide blood donation again. Used algorithms were compared by computing their classification performances.Publication Multicellular 4G and load balancing over cloud computing(Wiley, 2021-06-06) Hamd, Ravyar Jasim; Ali Yahiya, Tara; Kırcı, Pınar; KIRCI, PINAR; CZK-0182-2022LTE-based Multicellular 4G is nowadays gaining an enormous success thanks to its flexible core network features and its large data rate. As the number of users increases, along with the different services provided by the telcos, it is becoming a challenging issue to cope with the contextual increase in the backhaul traffic size: a congestion problem. In this perspective, when combining multicellular LTE with Cloud Computing (CC), this would increase the core network load and the subsequent load on the servers located in the CC that delivers services to the LTE users. To address this issue, we introduce an architecture combining both multicellular LTE and CC, and we investigate the performance of different load balancer mechanisms, namely static and semi-dynamic. Extensive simulation scenarios using different algorithms were set up. The obtained results were analyzed considering the Quality of Service (QoS) and computational resources for a deployed Web application.Publication Detection and separation of transparent objects from recyclable materials with sensors(Springer International Publishing Ag, 2022-01-01) Sentürk, Saruhan Furkan; Gülmez, Hüseyin Kaan; Gül, Mehmet Faruk; Kırcı, Pınar; Woungang, I.; Dhurandher, S. K.; Pattanaik, K. K.; Verma, A.; Verma, P.; Şentürk, Saruhan Furkan; Gülmez, Hüseyin Kaan; Gül, Mehmet Faruk; KIRCI, PINAR; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/Bilgisayar Mühendisliği Bölümü.; Woungang, I; Dhurandher, SK; Pattanaik, KK; Verma, A; Verma, P; FXP-2389-2022; EZQ-1139-2022; ICN-9200-2023; CZK-0182-2022Environmental pollution poses a threat to the ecological balance, economy and budgets of states. For this reason, many studies have been carried out recently to reduce environmental pollution, increase recycling, protect the environmental order and prevent economic deterioration. Many municipalities all over the world put recycling bins at certain points in cities, districts and even neighborhoods. However, these boxes are only designed to dispose of materials that can be recycled according to certain chambers with human hands and eyes. People separate their garbage and come and throw their garbage into that bin. Despite this, it can be seen that different wastes are in different bin and a manpower is needed to separate them again. In the project, recyclable materials (metals, transparent materials and others) will be separated into different bins by using sensors and smart systems without using human power.Publication A novel approach for monitoring of smart greenhouse and flowerpot parameters and detection of plant growth with sensors(MDPI, 2022-10-01) Kırcı, Pınar; Öztürk, Erdinç; Çelik, Yavuz; KIRCI, PINAR; Öztürk, Erdinç; Çelik, Yavuz; Bursa Uludağ Üniversitesi/Mühendislik Fakültesi/Bilgisayar Mühendisliği Bölümü; 0000-0003-1553-2619; E-5622-2010; CZK-0182-2022; JME-6512-2023Studies carried out in different parts of the world and in different climatic conditions have made it clear that it is very important to use smart technologies in solving the problems experienced in the field of agriculture globally and locally. Another important aim of the studies is to ensure that agricultural products are grown in smart greenhouse environments outside of arable lands. For this reason, growing agricultural products in greenhouses controlled by smart systems by creating suitable soil and climatic conditions and facilitating people's access to these products has been an important research and application subject. In this paper, the topics of "Smart Agriculture" and "Smart Greenhouse" were worked on. Therefore, a prototype of a smart greenhouse was constructed. Then, it was programmed according to the decided climatic conditions. Consequently, the main aim of this study was to improve the project according to the collected data by the sensors. One of the most important aims of our study was to question the possibility of growing different plants in the same greenhouse. In this context, in our study, a flower and three different vegetables with close growing conditions were grown in the same greenhouse in the same environment. These plants were grown individually both in the smart greenhouse prototype and in the pots outdoors in a natural environment. The differences between the two environments and the differences in the development of the plants were examined and the necessary results were obtained based on the findings. Based on the results obtained, it has been discussed what can be done if the plants grown in the greenhouse, of which we have created a small-scale prototype, are grown in a large-scale smart greenhouse. According to the results obtained, the smart greenhouse made a positive difference in the development of begonia, tomato and pepper. Although, the cucumbers grew more in the pots. In the study, it was observed that the plants were healthier in the smart greenhouse. The cucumbers in the greenhouse grew to 132 mm, the peppers to 61 mm and the tomatoes to 70 mm. The cucumbers in the pot grew to 163 mm, the peppers to 37 mm and the tomatoes to 60 mm. This shows that the yield was positively affected in the smart greenhouse. According to the collected results, the smart greenhouse system saved approximately 16.5% of water compared to the pot. The fact that the system can work both manually and autonomously provides a great convenience for the person controlling the greenhouse.