Shevchenko M.V.

Monitoring of Welding Processes with Application of Artificial Neural Networks

The paper presents a summary of methods of monitoring systems’ development for the processes involving heating of filler material and/ or base metal by the electric current and with periodical shortages of the welding circuit. The processes investigated were MAG welding, underwater flux-cored welding and flash-butt welding. Details of experiments, primary data processing procedures based on statistical analysis methods are described, the aim of primary processing being obtaining of informative parameters of the welding processes.

Mathematical Model of Arc Welding with Short Circuits for Development of Quality Monitoring System

The paper analyzes the influence of basic parameters of electric arc MAG welding – welding current, arc voltage, electrode diameter and outlet – on electrical processes in welding circuit. Also, we determine basic time parameters of electrode metal transfer for different transfer modes as well as their relationship with basic welding parameters. Furthermore, we develop the stochastic dynamical mathematical model of the welding process. The model includes electrical, physical and welding characteristics of equipment, preset values of welding parameters and electrode metal transfer mode.

Quality Monitoring of Welding by Electric Characteristics of the Process

This article presents the analysis results of existing methods of monitoring the quality of welded joints by electric parameters of the arc. Their advantages and disadvantages are described. We show that there is a need to use a large number of related parameters when creating automatic systems for monitoring the quality of welded joints by electrical parameters of the process. It requires a complex math-based environment. The systems of artificial intellect can be used as such an environment.

On Determining the Relation between Electrode Melting Speed and Impulse Parameters by Mig Welding with Modulated Current

The paper proposes the mathematical model for numerical solving of multi-parameter non-linear computational problem for analysis of the electrode melting process in the "electrode–arc" system. This model can be utilized under the conditions of pulsed metal arc welding by using argon or argon-based mixtures as a shielding gas. In addition, impulse parameters are calculated for the pulsed MIG welding. As a result, we obtain an area of impulse parameters which ensures the equality of electrode feed speed to its melting speed.

The quality monitoring method of welding power sources for arc welding using the artificial neural networks

In this paper, we considered the quality monitoring of welding power sources, comparing their static and dynamic behaviours with the standard behaviours. On the experimental side, we performed the troubleshooting of power sources with the help of the synthetic neural networks. We investigated the possibility of technical conditions evaluation of welding power sources.