Yazar "Turan, Sertan" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    A study on the monitoring of weld quality using XGBoost with Particle Swarm Optimization
    (Elsevier, 2024) Avci, Adem; Kocakulak, Mustafa; Acir, Nurettin; Gunes, Emrah; Turan, Sertan
    Gas Metal Arc Welding is a joining technique with numerous uses in manufacturing. Since welding process parameters have a considerable impact on the welding quality, online monitoring systems are utilized on production lines to achieve standard welding quality with minimum welding faults. This study presents preliminary work to develop a monitoring system for defect analysis in a Gas Metal Arc Welding process. This study aims to eliminate the need for laboratory tests with a model to reduce welding quality control costs. In this study, welding data such as voltage, current, and wire feeding rate were collected during the welding process in realtime. New features were derived from the gathered data at the preprocessing stage using statistical approaches. The determination of whether the welding process is defective or flawless was made using the Extreme Gradient Boosting Algorithm. The hyperparameters of the algorithm were optimized with Particle Swarm Optimization. The accuracy value was obtained as 93.15% after repeating the conducted experiments ten times. The recall, precision, specificity, and F1 -Score values in these experiments were calculated as 97.22%, 94.75%, 72.35%, and 95.94%, respectively. The mean current value was found to be the most relevant and meaningful feature that describes the welding quality based on intensive experiments. In addition to the proposed algorithm, some other machine -learning algorithms were tested on the welding dataset. With this study, the significance of feature derivation from the acquired welding current data has been discovered to analyze welding quality.
  • Küçük Resim Yok
    Öğe
    Front seat development for autonomous driving: A case of innovative product development
    (Gazi Üniversitesi, 2022) Altun, Koray; Berber, Reyhan Özcan; Kurt, Recep; Bektaş, Enes; Turan, Sertan; Korkmaz, Varol
    Automotive trends and the strategies of automotive OEMs (Original Equipment Manufacturers) indicate that advanced and fully autonomous vehicles will appear in the market soon. In parallel with these progresses, customers’ needs and expectations for automotive parts will evolve accordingly. Particularly, since the seat is in physical contact with the user, it requires major improvements by employing novel functions to meet these evolved needs and expectations. This study addresses these novel front seat functions and provides a real case of development of an innovative front seat for autonomous driving. In this study, initially, new design ideas were generated regarding the front seats to meet changing customer needs. Generated ideas were evaluated through the QFD methodology. And, in line with these selected ideas, the style design, CAD design, and finite element analysis and testing activities were carried out. As a result of this study, an innovative front seat for autonomous driving has been developed, by taking the evolved customer expectations into account, and by verifying the design through finite element analysis and tests. Consequently, evolving customer needs for front seat by autonomous driving are assessed and a real innovative product development case is presented in this study. Future work can address other relevant interior parts of the automotive as to be appropriate for autonomous driving.
  • Küçük Resim Yok
    Öğe
    Quality Determination by Using Support Vector Machine in Gas Welding Applications
    (Ieee, 2019) Avcı, Adem; Acır, Nurettin; Gunes, Emrah; Turan, Sertan
    The robots used in the manufacturing industry and the sensors from the automation system can be used to automatically perform quality checks. Gas welding robots can operate autonomously, but quality controls are carried out manually by means of laboratory tests. In this study, a method which can work fast in real time quality control applications is proposed by using the data obtained from the robots used in the production system. In this study, comparison of other classification algorithms which can be used in this field has been made. First of all, sensor data on the robots and production system were taken and quality control of the product at the end of the process was made and the entire process was classified. The processes in the obtained data were analyzed as raw data and statistical values were examined. Support Vector Machines, Decision Trees, Random Forests and Logistic Regression algorithms are used to classify the data. The algorithms used in the data set were successfully applied and a success rate of 87% was obtained with the Support Vector Machines.