USING THE METHOD OF CONFIRMATORY FACTOR ANALYSIS FOR THE SIMULATION OF TECHNICAL SYSTEMS

Abstract

Problem statement. Analyzing the dynamics of the functioning of complex technical systems, which are characterized by a certain architecture and the interaction of system components, in most cases, there are certain difficulties associated with the description of system-wide issues. Among such issues, first of all, it is possible to include problems related to the direct mathematical formalization of the overall system structure, the organization of relationships between its elements, the interaction of system elements with its external environment, control of the activities of its elements, etc. In such cases, most often researchers use the procedure of mathematical modeling. At the same time, there is a fairly large range of phenomena, in the analysis of which it is possible to use the traditional apparatus of mathematical modeling. In such situations, it is possible to apply modeling techniques that represent the model in the form of an algorithmic program for an electronic computer, the so-called simulation modeling. The purpose of the article. Application of simulated physical and mathematical modeling by the method of confirmatory factor analysis for the study of complex technical processes. Conclusion . The relationship between the parameters of the laser welding mode and the morphological criteria of the structural state of low-carbon low-alloy steel 09Г2С was investigated using the mathematical apparatus of simulation modeling, namely confirmatory factor analysis. The confirmatory factor method was implemented in the form of a path diagram, namely a graphical interpretation of the relationship between the structural condition criteria and laser welding parameters. The percentage content and geometric dimensions of the main structural components of ferrite and pearlite were used as criteria for the morphology of the structural state, and the geometric dimensions of the welded joint sections were used as welding parameters. The adequacy of the obtained model was confirmed by constructing a probability plot of normalized residuals according to the quasi-Newtonian method of residuals.