Development of machine learning method of titanium alloy properties identification in additive technologies
Journal Title: Восточно-Европейский журнал передовых технологий - Year 2018, Vol 3, Issue 12
Abstract
<p>Based on the experimentally established data on the parameters of microstructure, elemental and fractional composition of titanium alloy powders, four classes of their conformity (a material with excellent properties, optimal properties, possible defects in the material and defective material) as source raw materials for the additive technologies are identified. The basic characteristics of the material, which determine its belonging to a certain class, are established. Training and test samples based on 20 features that characterize each of the four classes of titanium alloy powders for the implementation of machine learning procedures were built. The developed method for identification of the class of material, based on the use of the second-order Kolmogorov-Gabor polynomial and the Random Forest algorithm, is described. An experimental comparison of the developed method work results with existing methods: Random Forest, Logistic Regression, and Support Vectors Machines based on the accuracy of their work in the training and application modes was made. The visualization of the results of all the investigated methods was given.</p><p>The developed supervised learning method allows constructing models for processing a large number of each input vector characteristics. In this case, the Random Forest algorithm provides satisfactory generalization properties while retaining the advantages of an additional increase of the accuracy based on the Kolmogorov-Gabor polynomial.</p><p>The main advantages of the developed method, in particular, regarding the additional increase of the accuracy of the classification task solution, are experimentally determined. The developed method allows increasing the modeling accuracy by 34.38, 33.34 and 3.13 % compared with the methods: Support Vectors Machine, Logistic Regression, and Random Forest respectively.</p><p>The obtained results allow one to considerably reduce financial and time expenses during the manufacture of products by additive technologies methods. The use of artificial intelligence tools can reduce the complexity and energy efficiency of experiments to determine the optimum characteristics of powder materials.<strong></strong></p>
Authors and Affiliations
Roman Tkachenko, Zoia Duriagina, Ihor Lemishka, Ivan Izonin, Andriy Trostianchyn
Keywords
Related Articles
- EP ID EP528039
- DOI 10.15587/1729-4061.2018.134319
- Views 56
- Downloads 0
Study of low-emission multi-component cements with a high content of supplementary cementitious materials
<p>The studies have established the influence of various types of supplementary cementitious materials on physical and mechanical properties and structure formation of low-emission multi-component cements. The results of...
Formalization of the concept of adaptive tasks mapping in the reconfigurable computers on FPGA
<p>The effectiveness of data processing in the reconfigurable computer systems depends significantly on the unproductive time costs of the reconfiguration of the FPGA computational space. It is an important modern proble...
Numerical methods for contact analysis of complex-shaped bodies with account for non-linear interface layers
<p>In order to ensure high technical characteristics of machines for various applications, it is necessary to increase the strength of the most loaded and heavy-duty elements of constructions, which are complex-shaped co...
Uncertainty reduction in big data catalogue for information product quality evaluation
<p class="Literature">The method of vitality of the information product evaluation is built. This method, as opposed to others, takes into account components of the information product, which gave an opportunity to forec...
Modification of the PERT method for project time evaluation taking into account unexpected delays
<p>The PERT-based method for estimation of project implementation time, which uses Rayleigh distribution instead of β-distribution, was proposed. The modification of the PERT, which made it possible to involve the qualit...