Durmus Karayel

Simulation of Direct Extrusion Process and Optimal Design of Technological Parameters Using FEM and Artificial Neural Network

This study aims the modeling and simulation of the direct extrusion process and determination of optimal process parameters using Finite Element Method (FEM) and Artificial Neural Network (ANN) cooperatively. First, the die set has been designed for direct extrusion of an aluminum rod and its numerical simulation has been prepared by mean of ABAQUS/EXPLITIC finite element code. So, both the values of the process parameters according to extrusion load and the critical stress values have been determined. After, the ANN model of the process has been developed under MATLAB and has been trained with the results of finite element simulations. Also, the optimization software which can run together the ANN model has been developed and has been used to determine the optimum process parameters.

Modeling and Analysis of the Weld Bead Geometry in Submerged Arc Welding by Using Adaptive Neurofuzzy Inference System

This study is aimed at obtaining a relationship between the values defining bead geometry and the welding parameters and also to select optimum welding parameters. For this reason, an experimental study has been realized. The welding parameters such as the arc current, arc voltage, and welding speed which have the most effect on bead geometry are considered, and the other parameters are held as constant. Four, three, and five different values for the arc current, the arc voltage, and welding speed are used, respectively. So, sixty samples made of St 52-3 material were prepared. The bead geometries of the samples are analyzed, and the thickness and penetration values of the weld bead are measured. Then, the relationship between the welding parameters is modeled by using artificial neural network (ANN) and neurofuzzy system approach. Each model is checked for its adequacy by using test data which are selected from experimental results. Then, the models developed are compared with regard to accuracy. Also, the appropriate welding parameters values can be easily selected when the models improve.

Design and Implementation of the Control System of an Internal Combustion Engine Test Unit

Accurate tests and performance analysis of engines are required to minimize measurement errors and so the use of the advanced test equipment is imperative. In other words, the reliable test results depend on the measurement of many parameters and recording the experimental data accurately which is depended on engine test unit. This study aims to design the control system of an internal combustion engine test unit. In the study, the performance parameters of an available internal combustion engine have been transferred to computer in real time. A data acquisition (DAQ) card has been used to transfer the experimental data to the computer. Also, a user interface has been developed for performing the necessary procedures by using LabVIEW. The dynamometer load, the fuel consumption, and the desired speed can easily be adjusted precisely by using DAQ card and the user interface during the engine test. Load, fuel consumption, and temperature values (the engine inlet-outlet, exhaust inlet-outlet, oil, and environment) can be seen on the interface and also these values can be recorded to the computer. It is expected that developed system will contribute both to the education of students and to the researchers’ studies and so it will eliminate a major lack.

Integrated Knowledge-Based System for Machine Design

An integrated design system (IDS) approach has been developed to integrate various stages of the mechanical design process, including rapid prototyping. The system consist, of design, analysis, calculation, rapid prototyping, and library modules, and it blends artificial intelligence methods, CAD-CAM, and technical computing packages into a single environment. The system has been applied for the design of one-stage gearbox with helical gear. In this case study, all stages of the design process are carried out by using IDS.

Esnek Algılayıcı Kontrollü Robot El Tasarımı ve Gerçeklenmesi

Dunya genelinde artan yalin uretim sistemlerinin gelismesinde suphesiz en onemli degere sahip olan girdilerden birisi de robotik calismalardir. Robotik calismalar sayesinde uretimde olusabilecek hata paylari ve kaza oranlari en aza indirilerek uretimler gerceklestirilebilmektedir. Endustriyel yapidaki robotlarin insansi robotlar (humanoid) uzerine yogunlasmasini takiben uretimde hassasiyet ve kalite her gecen gun artmaktadir. Bu yaklasimla, robot manipulatorlerinin zamanla insan eli yapisina olan benzerliklerinin artmasi gozlemlenmektedir. Bu calismada, eldivene monte edilmis esneklik sensorleri yardimiyla insan elinden alinan konum bilgileri ile mekatronik tabanli robotik elinin kontrolu saglanmistir. Insan el parmaklarinin acisal hareketlerini algilanarak bir mikrodenetleyici tarafindan islendikten sonra servo motorlar yardimiyla robot elin kontrolu saglanmistir. Calisma sonunda ortaya cikan robot el ile insan icin tehlikeli ortamlarda manipulasyon islemlerini uzaktan kontrol etmek amaclanmistir.

Development of a wearable exercise device for rehabilitation of hemiplegic hand

Millions of people suffer a stroke each year. One of the problems after stroke is hemiplegic hand. In previous studies a lot of robotic based devices has been suggested for rehabilitation of hemiplegic hand. Actuators using for force transmitting in previous devices have some disadvantages. One of them is misalignment on the centre of rotation between joints used in devices and joints of the hand. In this study a novel device is suggested to perform the finger exercises. The device has a cable and spring driven mechanism with an electrical linear actuator. The device allows both passive and active exercises. It is aimed that the device is suitable for at home use, wearable, portable and low cost. It is expected from the new device that it will reduces the rehabilitation cost and therapy period. The device can be used for other rehabilitation treatments such as nerve injury, nerve compression, tendon injury, fracture and sport injury besides stroke rehabilitation.

Determination of Factors Affecting Safety Stress in Machine Design by Using Artificial Intelligence Technologies

In this study, an intelligent system model that can evaluate experimental material properties and safety factors is developed. The model contains Artificial Intelligence Technologies such as Artificial Neural Network (ANN) and Fuzzy Logic. It consists of sub modules into interaction. Also, the model can obtain more precision values than interpolation techniques used to classical design. The study contributes to define safety factors, design criterions and safety stress according to a new approach based on information technologies. So, this study can be seen as one of the sub modules of Intelligence Multi Agent System and it can be integrated with Multi Agent System Model for design. Also, it can be used for classical design studies so that results can be quickly obtained. It is expected that this approach will be widely used by designers.