Mathematical model for hardness test for steel R6M5

Abstract

Introduction. The evaluation of the quality criteria of the instrumental quick-cutting steel R6M5 by non-destructive methods is an urgent task. This is due to the fact that tools made from it are used to work on cutting, grinding, drilling machines, since it is not always technically possible to carry out full-scale tests on such parts. The paper proposes a method for evaluating the hardness of steel with the use of mathematical modelling. Materials and methods. In this work the influence of elements of the chemical composition of steel R6М5 on the hardness indexes was investigated using the method of experiment planning. The chemical composition of the steel changed in accordance with accepted regulatory documents (GOST 19265). Experiment results. The working area of the selected parameters (indicators of elements of chemical composition) and the objective function (hardness) were investigated. The range of numerical values of the parameters of the working area was limited to the limiting values of the indicators of the chemical composition of steel R6M5. To build the planning matrix of experiments, only those lines were selected for which information on hardness indicators was known from literary sources or based on the analysis of expert information. In the process of implementation of the matrix of experimental planning, a regression model for evaluating the hardness values of steel R6M5, depending on its chemical composition, was obtained. The model is adequate according to Fisher criteria (F = 1,002) and Kochren (F = 0,336). Based on the analysis of the model, the histogram of the influence of the elements of the chemical composition on the hardness indexes is constructed, which is confirmed by their physicochemical interaction. It was found that the greatest influence on the hardness of steel R6M5 is given by carbon (1,050), chromium (0,550) and tungsten (0,275). Conclusions. For steel R6М5 by the method of planning of experiments the model of the prediction of hardness parameters depending on the chemical composition is obtained, which leads to save the material and time costs for natural tests.