WSEAS Transactions on Environment and Development
Print ISSN: 1790-5079, E-ISSN: 2224-3496
Volume 15, 2019
Application of Artificial Neural Networks for Prediction of Nickel-based Superalloys Service Properties based on the Chemical Composition
Authors: , ,
Abstract: Nickel-based superalloys are utilized in various fronts of industry; however, they are especially important in turbine building. To date, a large amount of data on the chemical compositions and performance of the alloys has been accumulated; however, usually, there is only data on a rather narrow range of temperatures and test exposures. The lack of information on the properties of alloys outside the studied ranges is a significant limitation on their possible applications. The available approaches are based mainly on practical experiments and, therefore, are expensive and time consuming. Moreover, the alloys development has a significant environmental impact since the chemical composition compound is conjugated with emissions and disposals of toxic and radioactive heavy metals. Up to date, it is possible to substantiate a new approach to analyzing the service properties of metallic materials based on the use of artificial neural networks. The previously obtained real test data formed the basis for estimation the missing values. In this work, we have studied the database of 210 nickel-based superalloys with regard to their tensile strengths in various conditions. The chemical composition and test conditions of the alloys were the input parameters of the ANN. Based on the ANN forecast, the dependences of tensile strength on temperature and time conditions were obtained, which were presented in the form of Larson-Miller parameter. The analysis of dependences allows one to select the parameters of the initial heat resistance (tensile strength) and thermal stability of the alloys. In addition, the dependence of tensile strength and thermal stability on the content of the main reinforcing elements was obtained using ANN modeling. The calculation results coincide with the experimental data.
Search Articles
Keywords: nickel-based superalloys, artificial neural network, tensile strength, heat resistance, thermal stability
Pages: 113-119
WSEAS Transactions on Environment and Development, ISSN / E-ISSN: 1790-5079 / 2224-3496, Volume 15, 2019, Art. #13