Intellectual Control of Induction Soldering Process using Neuro-fuzzy Controller

Abstract

The article presents an approach to solving the task of controlling the technological process of induction soldering of thin-walled aluminum waveguide paths using a neuro-fuzzy controller. The paper provides an analysis of the subject area, and also articulates the task of managing the technological process of induction soldering based on intelligent methods. The selection and comparison of intelligent methods of process control is made. An experimental study of the proposed methods effectiveness showed that the method of controlling the technological process of waveguide paths induction soldering based on a neuro-fuzzy controller shows the greatest efficiency. Based on this approach and using the Python programming language a control module for the existing automated control system for induction soldering was made. A block diagram of the module algorithm, as well as a description of its work is shown. The approbation of the proposed approach was carried out when soldering several pipe sizes of waveguide paths: 58 – 25 mm, 35 – 15 mm and 19 – 9.5 mm. In all cases, high-quality soldered connections of waveguide paths were obtained.