ABSTRACT This study conducts failure analysis of filament wound glass reinforced plastic (GRP) pi... more ABSTRACT This study conducts failure analysis of filament wound glass reinforced plastic (GRP) pipes made of E-glass/epoxy with an inclined surface crack under open-ended internal pressure. Test specimens are composed of six antisymmetric layers with (±55°)3 winding angles. Tests have been performed at seven different crack angles: 0°,15°,30°,60°,45°,75°, and 90°, with crack-to-thickness ratio of a/t=0.50. The burst strengths of the specimens were determined, and the dependence of the burst strength on the crack angle was examined.
The current automotive industry uses new technology in materials processing and lightweight mater... more The current automotive industry uses new technology in materials processing and lightweight materials to minimize vehicle weight and fuel consumption. As a result, Tailor-Welded Blanks (TWB) are used in regions where additional strength or stiffness is necessary in body-in-white, thus decreasing weight. In this paper, TWB were produced by joining DP600 with 1.8 mm and DP800 with 0.8 mm and 1.5 mm using CO 2 laser welding. The effects of laser power and welding speed on the mechanical properties of TWB were investigated. The flow behaviors of the TWB were obtained from tensile tests. The microstructures of the weld zones of the base materials and TWB were revealed, and the numerical analyzes of the microstructures were performed by micromechanical modeling with a 2D representative volume element (RVE). Effective stress and plastic strain distributions in the microstructures and macroscopic flow curves of the base materials and the weld zones were obtained. The uniaxial tensile tests of the samples were modeled, and the flow behaviors of the TWB were obtained with micromechanical modeling. The results showed a significant change in stress distribution and strain localization depending on the grain size of constituents and the thickness ratios of the TWB at micro and macro scales, respectively. The numerical flow curves obtained using the RVE method were in good agreement with those of the experimental.
In this paper, we use Kinect device to detect human hand movements and we explain how we can buil... more In this paper, we use Kinect device to detect human hand movements and we explain how we can build a city with human hand movements. Kinect is a system that can detect human movements and send them to computers. It was developed by Microsoft to play games for the Xbox game console and is being used over time for applications in other areas. Because of the new technology, there is a lack of application and the literature in this field. In this study , the user's hand movements were detected by Kinect and is done the modeling of the city by drag and drop method by accessing shapes that was designed in the WPF screen.
In this study , uni-directionaly reinforced laminates that arranged e-glass / epoxy composite sam... more In this study , uni-directionaly reinforced laminates that arranged e-glass / epoxy composite samples in the form [ + 45 / -45 / 90/0 ] s and [ + 45 / -45 / 90/0 ] 2s was applied low-velocity impact tests at impact energies of 10, 20 and 30 J has been obtained 8 and 16 laminated sample. Low-velocity impact tests were performed with the drop weight test device. Low-velocity impact damage was created on the samples depending on the different impact energies as a result of experiments conducted with hemispherical various impactor. The impactor mass was 6.350 kg. Digital imaging process method was applied images of the samples subjected to impact damage and damage maps for each damage zone were extracted. Also, change graph of damage area depending on the impact energies has been omitted. Changing damage zones were extracted. Also, change graph of damage area depending on the impact energies has been omitted.
Artificial neural network (ANN) may achieve successful classification value, but obtained results... more Artificial neural network (ANN) may achieve successful classification value, but obtained results sometimes can be unintelligible and may not be interpreted. In literature, different rule extraction methods from trained ANN are applied to overcome this problem. This study presented a new method of extraction correct and intelligible rules from Trained ANN using variable neighbourhood search (VNS) metaheuristic method. Since the VNS method is independent from the ANN, it does not modify the results of ANN. Real time Wisconsin Breast Cancer (WBC) data was examined for determining the feasibility of the suggested method. The VNS method was used to obtain the best fitness values belong to input attributes, Id, which was maximized the fitness function Sr of output node r. The suggested method was the computational evaluated on investigated data sets and obtained results from the suggested approach indicated that had a potential to produce accurately and cored rules. The proposed method o...
International Advanced Researches and Engineering Journal, 2021
In this study, the static, kinematic and dynamic behaviors of a semi-active hand prosthesis were ... more In this study, the static, kinematic and dynamic behaviors of a semi-active hand prosthesis were analyzed numerically. Finite Elements method was used in static analysis and analytical method was used in kinematic and dynamic analysis. The mathematical model of the hand was created in kinematic and dynamic analysis. Using the mathematical model obtained, torque values of 0, 15, 30, 45, 60, 75, 90 degrees were calculated according to the different position angles of the fingers. Examination was performed for 4 fingers (index finger, middle finger, ring finger, little finger) and 5 kg of force was applied to the fingertips perpendicular to the finger plane. In this examination, the forces are divided into 25% for the index finger, 35% for the middle finger, 25% for the ring finger and 15% for the little finger. The results obtained for forces at different angles under the specified conditions were explained. As a result of all these stages, the prosthetic hand was designed. The design...
Artificial neural network (ANN) may achieve successful classification value, but obtained results... more Artificial neural network (ANN) may achieve successful classification value, but obtained results sometimes can be unintelligible and may not be interpreted. In literature, different rule extraction methods from trained ANN are applied to overcome this problem. This study presented a new method of extraction correct and intelligible rules from Trained ANN using variable neighbourhood search (VNS) metaheuristic method. Since the VNS method is independent from the ANN, it does not modify the results of ANN. Real time Wisconsin Breast Cancer (WBC) data was examined for determining the feasibility of the suggested method. The VNS method was used to obtain the best fitness values belong to input attributes, Id, which was maximized the fitness function Sr of output node r. The suggested method was the computational evaluated on investigated data sets and obtained results from the suggested approach indicated that had a potential to produce accurately and cored rules. The proposed method o...
Journal of Reinforced Plastics and Composites, 2005
In this study, mechanical buckling behaviors of composite square plates of four symmetric and ant... more In this study, mechanical buckling behaviors of composite square plates of four symmetric and antisymmetric layers with holes are examined. A thermoplastic material is used as the matrix material. Two different reinforced materials are used, of which one is a galvanized steel wire and the other a woven galvanized wire. The hole type is taken as circular. The finite element method is used for the solution. First critical buckling loads of an orthotropic plate with and without hole subjected to unidirectional compression are found. The results are illustrated in figures.
In this study, mechanical buckling behaviors of composite square plates of four symmetric and ant... more In this study, mechanical buckling behaviors of composite square plates of four symmetric and antisymmetric layers with holes are examined. A thermoplastic material is used as the matrix material. Two different reinforced materials are used, of which one is a galvanized steel wire and the other a woven galvanized wire. The hole type is taken as circular. The finite element method is used for the solution. First critical buckling loads of an orthotropic plate with and without hole subjected to unidirectional compression are found. The results are illustrated in figures.
Artificial neural network (ANN) may achieve successful classification value, but obtained results... more Artificial neural network (ANN) may achieve successful classification value, but obtained results sometimes can be unintelligible and may not be interpreted. In literature, different rule extraction methods from trained ANN are applied to overcome this problem. This study presented a new method of extraction correct and intelligible rules from Trained ANN using variable neighbourhood search (VNS) metaheuristic method. Since the VNS method is independent from the ANN, it does not modify the results of ANN. Real time Wisconsin Breast Cancer (WBC) data was examined for determining the feasibility of the suggested method. The VNS method was used to obtain the best fitness values belong to input attributes, Id, which was maximized the fitness function Sr of output node r. The suggested method was the computational evaluated on investigated data sets and obtained results from the suggested approach indicated that had a potential to produce accurately and cored rules. The proposed method o...
ABSTRACT This study conducts failure analysis of filament wound glass reinforced plastic (GRP) pi... more ABSTRACT This study conducts failure analysis of filament wound glass reinforced plastic (GRP) pipes made of E-glass/epoxy with an inclined surface crack under open-ended internal pressure. Test specimens are composed of six antisymmetric layers with (±55°)3 winding angles. Tests have been performed at seven different crack angles: 0°,15°,30°,60°,45°,75°, and 90°, with crack-to-thickness ratio of a/t=0.50. The burst strengths of the specimens were determined, and the dependence of the burst strength on the crack angle was examined.
The current automotive industry uses new technology in materials processing and lightweight mater... more The current automotive industry uses new technology in materials processing and lightweight materials to minimize vehicle weight and fuel consumption. As a result, Tailor-Welded Blanks (TWB) are used in regions where additional strength or stiffness is necessary in body-in-white, thus decreasing weight. In this paper, TWB were produced by joining DP600 with 1.8 mm and DP800 with 0.8 mm and 1.5 mm using CO 2 laser welding. The effects of laser power and welding speed on the mechanical properties of TWB were investigated. The flow behaviors of the TWB were obtained from tensile tests. The microstructures of the weld zones of the base materials and TWB were revealed, and the numerical analyzes of the microstructures were performed by micromechanical modeling with a 2D representative volume element (RVE). Effective stress and plastic strain distributions in the microstructures and macroscopic flow curves of the base materials and the weld zones were obtained. The uniaxial tensile tests of the samples were modeled, and the flow behaviors of the TWB were obtained with micromechanical modeling. The results showed a significant change in stress distribution and strain localization depending on the grain size of constituents and the thickness ratios of the TWB at micro and macro scales, respectively. The numerical flow curves obtained using the RVE method were in good agreement with those of the experimental.
In this paper, we use Kinect device to detect human hand movements and we explain how we can buil... more In this paper, we use Kinect device to detect human hand movements and we explain how we can build a city with human hand movements. Kinect is a system that can detect human movements and send them to computers. It was developed by Microsoft to play games for the Xbox game console and is being used over time for applications in other areas. Because of the new technology, there is a lack of application and the literature in this field. In this study , the user's hand movements were detected by Kinect and is done the modeling of the city by drag and drop method by accessing shapes that was designed in the WPF screen.
In this study , uni-directionaly reinforced laminates that arranged e-glass / epoxy composite sam... more In this study , uni-directionaly reinforced laminates that arranged e-glass / epoxy composite samples in the form [ + 45 / -45 / 90/0 ] s and [ + 45 / -45 / 90/0 ] 2s was applied low-velocity impact tests at impact energies of 10, 20 and 30 J has been obtained 8 and 16 laminated sample. Low-velocity impact tests were performed with the drop weight test device. Low-velocity impact damage was created on the samples depending on the different impact energies as a result of experiments conducted with hemispherical various impactor. The impactor mass was 6.350 kg. Digital imaging process method was applied images of the samples subjected to impact damage and damage maps for each damage zone were extracted. Also, change graph of damage area depending on the impact energies has been omitted. Changing damage zones were extracted. Also, change graph of damage area depending on the impact energies has been omitted.
Artificial neural network (ANN) may achieve successful classification value, but obtained results... more Artificial neural network (ANN) may achieve successful classification value, but obtained results sometimes can be unintelligible and may not be interpreted. In literature, different rule extraction methods from trained ANN are applied to overcome this problem. This study presented a new method of extraction correct and intelligible rules from Trained ANN using variable neighbourhood search (VNS) metaheuristic method. Since the VNS method is independent from the ANN, it does not modify the results of ANN. Real time Wisconsin Breast Cancer (WBC) data was examined for determining the feasibility of the suggested method. The VNS method was used to obtain the best fitness values belong to input attributes, Id, which was maximized the fitness function Sr of output node r. The suggested method was the computational evaluated on investigated data sets and obtained results from the suggested approach indicated that had a potential to produce accurately and cored rules. The proposed method o...
International Advanced Researches and Engineering Journal, 2021
In this study, the static, kinematic and dynamic behaviors of a semi-active hand prosthesis were ... more In this study, the static, kinematic and dynamic behaviors of a semi-active hand prosthesis were analyzed numerically. Finite Elements method was used in static analysis and analytical method was used in kinematic and dynamic analysis. The mathematical model of the hand was created in kinematic and dynamic analysis. Using the mathematical model obtained, torque values of 0, 15, 30, 45, 60, 75, 90 degrees were calculated according to the different position angles of the fingers. Examination was performed for 4 fingers (index finger, middle finger, ring finger, little finger) and 5 kg of force was applied to the fingertips perpendicular to the finger plane. In this examination, the forces are divided into 25% for the index finger, 35% for the middle finger, 25% for the ring finger and 15% for the little finger. The results obtained for forces at different angles under the specified conditions were explained. As a result of all these stages, the prosthetic hand was designed. The design...
Artificial neural network (ANN) may achieve successful classification value, but obtained results... more Artificial neural network (ANN) may achieve successful classification value, but obtained results sometimes can be unintelligible and may not be interpreted. In literature, different rule extraction methods from trained ANN are applied to overcome this problem. This study presented a new method of extraction correct and intelligible rules from Trained ANN using variable neighbourhood search (VNS) metaheuristic method. Since the VNS method is independent from the ANN, it does not modify the results of ANN. Real time Wisconsin Breast Cancer (WBC) data was examined for determining the feasibility of the suggested method. The VNS method was used to obtain the best fitness values belong to input attributes, Id, which was maximized the fitness function Sr of output node r. The suggested method was the computational evaluated on investigated data sets and obtained results from the suggested approach indicated that had a potential to produce accurately and cored rules. The proposed method o...
Journal of Reinforced Plastics and Composites, 2005
In this study, mechanical buckling behaviors of composite square plates of four symmetric and ant... more In this study, mechanical buckling behaviors of composite square plates of four symmetric and antisymmetric layers with holes are examined. A thermoplastic material is used as the matrix material. Two different reinforced materials are used, of which one is a galvanized steel wire and the other a woven galvanized wire. The hole type is taken as circular. The finite element method is used for the solution. First critical buckling loads of an orthotropic plate with and without hole subjected to unidirectional compression are found. The results are illustrated in figures.
In this study, mechanical buckling behaviors of composite square plates of four symmetric and ant... more In this study, mechanical buckling behaviors of composite square plates of four symmetric and antisymmetric layers with holes are examined. A thermoplastic material is used as the matrix material. Two different reinforced materials are used, of which one is a galvanized steel wire and the other a woven galvanized wire. The hole type is taken as circular. The finite element method is used for the solution. First critical buckling loads of an orthotropic plate with and without hole subjected to unidirectional compression are found. The results are illustrated in figures.
Artificial neural network (ANN) may achieve successful classification value, but obtained results... more Artificial neural network (ANN) may achieve successful classification value, but obtained results sometimes can be unintelligible and may not be interpreted. In literature, different rule extraction methods from trained ANN are applied to overcome this problem. This study presented a new method of extraction correct and intelligible rules from Trained ANN using variable neighbourhood search (VNS) metaheuristic method. Since the VNS method is independent from the ANN, it does not modify the results of ANN. Real time Wisconsin Breast Cancer (WBC) data was examined for determining the feasibility of the suggested method. The VNS method was used to obtain the best fitness values belong to input attributes, Id, which was maximized the fitness function Sr of output node r. The suggested method was the computational evaluated on investigated data sets and obtained results from the suggested approach indicated that had a potential to produce accurately and cored rules. The proposed method o...
Uploads