Naslov
Prediction of the nitriding parameters by neural network and genetic algorithm

Autori
Filetin, Tomislav ; Žmak, Irena ; Novak, Davor:

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Thermal Process Modelling and Computer Simulation (ICTPMCS) / - Nancy : Ecole des Mines de Nancy and Societe Francaise et de Materiaux, 2003, 37-37

Skup
2nd Int. Conference onT hermal Process Modelling and Computer Simulation

Mjesto i datum
Francuska, 31.03.2003. - 02.04.2003

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
nitriding; neural network; genetic algorithm; predicting of parameters

Sažetak
Neural network, genetic algorithm and genetic programming, as one of the attractive methods in artificial intelligence, have a great potential and possibilities to solve problems in complex nonlinear system modelling, estimating, predicting, the diagnosis and the adaptive control. Recently, this has become possible with the prediction of material properties and determination of heat treatment process parameters. The surface hardness and hardness profile of a nitrided workpiece depend on the chemical composition of the steel, nitriding temperature and time, and on type of the nitriding process (i.e. atmosphere). An issue in this approach was to test how the statistical analysis, artificial neural network, genetic algorithm and genetic programming may be used for determination of nitriding time and surface hardness, in case when the chemical composition of steel, nitriding temperature and required thickness of nitrided layer are known. In the neural network learning procedure datasets of results were used, after nitriding 5 different steels (chemical compositions) -42CrMo4, 31CrMoV9, X32CrMoV3-3, X40CrMoV5-1, X35CrMo17. Different combinations of time, temperature, surface hardness and thickness of plasma and gas nitriding layer are compiled from the experiments (B. Edenhofer, H. Trenkler) and industrial experiences and also from the literature. The static multi-layer feed-forward neural network is proposed. To accelerate the convergence of the proposed static error-back propagation learning algorithm, the momentum method is applied. The mean error between experimental data of nitriding time and data predicted by means of neural network, and also the standard deviation for the testing dataset are is small and acceptable. The determination of time by genetic algorithm gives a greater standard deviation then by neural network. The artificial neural network offers a simple and effective new tool to predict the nitriding time and surface hardness.

Izvorni jezik
Engleski

Znanstvena područja
Strojarstvo

POVEZANOST RADA