Akademik Veri Yönetim Sistemi
MATERIALS TESTING, cilt.61, sa.6, ss.567-572, 2019 (SCI-Expanded)
Since surface hardness affects wear resistance and case depth affects fatigue strength, the optimum value of both is extremely important with respect to the area of use. The aim of this study was to investigate the possibility of predicting case depth and surface hardness in ion nitrided AISI 4340 steels as a function of process time and temperature by using artificial neural networks and to obtain useful case depth and surface hardness data from an artificial neural networks model. Two projections were created for ion nitrided case depth and surface hardness, both depending on process time and temperature, and the conclusion was reached that the experimental data provides sufficient predictability regarding the artificial neural networks model . In the multilayer perceptron artificial neural networks architecture designed, two outputs (case depth and surface hardness) were determined in the same network according to the inputs, thus providing the integrity of the system characterization. The system was created by means of a Matlab simulink graphical user interface, which determined the artificial neural networks outputs according to the specified input with the purpose of visualizing the process. Different input values could be entered for visually determining the output values of the process.