Hüdayim Başak

Chain gear design using artificial neural networks

In this study, the analytical calculation and analysis for the data corresponding to these gears have been made to obtain alternative test and training data sets to be used at the artificial neural networks(ANNs) with the constraints and requirements for the design of chain gears of power and motion transmission mechanisms are determined. In the input layer, the constraints and requirement values(input power, number of revolution of the pinion, number of revolution of the gear and center distance) of chain gears are used while at the output layer chain code(i.e., the chain gear type and chain sequence number), specifying the functional and physical properties, is used. Then the network is tested with the test data. The analytical calculation results and ANN predictions are compared by using statistical error analyzing methods such as absolute fraction of variance(R2), root mean square error(RMSE), and mean error percentage(MEP) for the training and test data. The chain code has been determined by the ANN with an acceptable accuracy. It is concluded that ANNs can be used as an alternative method in chain gear design.

Comparing Microstructures and Tensile Properties of Intercritically Annealed and Quenched-Tempered 1.7Ni–1.5Cu–0.5Mo–0.2C Powder Metallurgy Steels

The aim of this study was to compare the influence of intercritical quenching (IQ), step quenching (SQ) and quenching plus tempering (QT) heat treatments on the microstructure and tensile properties of 1.7Ni–1.5Cu–0.5Mo–0.2C pre-alloyed powder metallurgy (P/M) steels. In the microstructures of the IQ and SQ specimens partial martensite having Ni-rich phases formed up in the soft ferritic matrix. It was observed that unlike Mo, a Cu alloying element dissolved homogeneously in the specimens. The martensite volume fraction (MVF) in the SQ specimens was higher than that in the IQ specimens. It was found that macrohardness, yield and tensile strengths increased, whereas microhardness of ferrite and elongation decreased with increasing MVF. However, with this increase, microhardness values of martensite phases decreased in the IQ specimen, while they increased in SQ specimens. It was observed that the yield, tensile, and elongation values of the QT specimens were lower than those of all intercritically annealed specimens having the same hardness values.

ISIL İŞLEM UYGULANMIŞ ORTA KARBONLU ÇELİKTE MİKROYAPI DEĞİŞİMİNİN BULANIK MANTIK İLE TAHMİN EDİLMESİ

Bu calismada, su verme islemi ile mikroyapisi martenzit faza getirilmis, AISI 1050 orta karbonlu celige uc farkli isil islem sicakligi ve uc farkli isil islem suresi uygulanmis ve bu islemler sonucunda mikroyapida gozlenen degisimler bulanik mantik ile tahmin edilmistir. Mikroyapi degisimlerini tahmin etmek icin uc farkli sicaklik 500 °C , 600 °C ve 700 °C ve uc farkli sure 15, 60 ve 180 dakika giris parametresi olarak secilmistir. Giris parametrelerine bagli olarak mikroyapinin kuresellesme orani dilsel ifade olarak hic, az, orta ve cok sozcukleri ile bulaniklastirilarak dokuz adet kural olusturulmustur. Kuresellestirme orani tahminlerini yapilabilmesi icin Mamdani cikarim yontemi kullanilmistir. Sonuc olarak bulanik mantik ile orta karbonlu celigin mikroyapi degisimleri basarili bir sekilde tahmin edildigi gorulmustur.

The Assessment of Cutting Force with Taguchi Design in Medium Carbon Steel–Applied Spheroidization Heat Treatment

In this study, the analysis of cutting forces on medium carbon AISI 1050—to which different spheroidization heat treatments were applied—was conducted by the mixed-level Taguchi orthogonal experiment design method. In the experiments, besides the parameters of feed rate, depth of cut and cutting speed having effect on cutting forces as a factor in orthogonal design, the spheroidization time and temperature parameters were also used. By the performed orthogonal experiment design method, the values of cutting forces were estimated using the five-factor, two- and three-leveled Taguchi L36 (22 × 33) mixed orthogonal experiment design method. The effectiveness of the machining parameters on the cutting force was revealed by performing analysis of variance test. Moreover, the effectiveness rates of the parameters were also determined in the study as per the signal noise rates. Consequently, it has been observed that the feed rate is more effective on the cutting forces compared to other parameters.

Küreselleştirme isil İşlemi uygulanmiş AISI 1050 Çeliğinin yüzey pürüzlülük değerlerinin yapay sinir ağlari ile modellenmesi

Estimation of surface roughness values, which is an indication of workpiece quality, is important in terms of reducing the cost and duration of machining. In this study, the surface roughness values of the medium carbon steel subjected to the spheronization heat treatment have estimated by artificial neural networks. ANN network model have been created by being chosen feedforward back propagation network model, the adoption of network structure and learning function LEARNGDM, TRAINLM as training algorithm, MSE for assessment of network performance and two hidden layers. The value of each neuron in the network have been transferred another layer by TANSIG, LOGSIG and PURELIN transfer functions. As a result, the artificial neural networks trained and tested have been found to be easy to use for estimating surface roughness values with a high percentage of R = 0.99001 according to MSE performance.

The mechanical strength of aluminum alloys which are joined with friction stir welding modelling with artificial neural networks

Artificial neural networks (ANN) is increasingly being used to solve engineering problems. Successful welding operations are at the top of engineering problems. The friction stir welding method, which is used to combine the materials under the melting temperature, is also a method that can be analyzed with ANN. In this study, Al-7075 workpieces (with and without aging treatment) were joined by the friction stir welding method. Tension tests were applied to the joined workpieces and the resulting values were modeled by ANN. In ANN model; The tensile strength of the assembled parts was tried to be estimated by input parameters (the type of the material, the number of revolutions and the feed rate). In the created ANN model; Different nerve numbers, different transfer functions, using different layer numbers, the models created by alternative network structures have been tested. In these experiments, prediction performance was measured using neuron and algorithm type variables. At the end of this study, it was seen that the ANN model for the parts joined with the FSW process made predictions with high consistency. The best estimates were obtained with the Levenberg Marquardt (LM) algorithm and the ANN model established with 3 neurons. This ANN model is seen as a suitable solution for this problem with 98% success.

Comprehensive gearbox design using expert system techniques

In this study, a computer aided design (CAD) software is developed in order to design various gearboxes under several conditions, to draw detailed assembly drawings and to carry out comprehensive gearbox analysis. An expert system computer language, VisualLISP, is used for this purpose. With the developed software, gear designs and shaft analysis can be carried out by using necessary parameters for the gear box. Certain detailed analyses such as rolling bearing selection and gearbox volume determination are realized by the expert system techniques.

Comprehensive design of rotating shafts and a software for designing and selecting of rolling‐contact bearings

In this study a computer program was developed using Visual Basic and Visual LISP programming languages to comprehensively analyze the strength, deformation, vibration, manufacturing, and drawing of a rotating circular stepped shaft. These shafts have a variety of power transmitting mechanical elements (components), such as pairs of gears, and pulleys that are modeled in 3D. In addition, force and moment analyses of these elements running on the shafts were examined. Notch sensitivity and stress concentration factors given graphically were formulated using the finite difference method. Vibration analysis was also included to the software. Furthermore, to select the best rolling‐contact bearings, an expert system requiring the knowledge‐based technique was employed. The user is given the choice of shaft material and other parameters. After saving the data regarding calculated and designed shaft, these data are transferred to the graphical environment for detail drawing and CNC code generation of the shaft. A program, which is called shaft and gear design program, has been developed for this process.