Publishing House SB RAS:

Publishing House SB RAS:

Address of the Publishing House SB RAS:
Morskoy pr. 2, 630090 Novosibirsk, Russia



Advanced Search

Avtometriya

2025 year, number 2

STUDY OF DROSS OCCURRENCE IN A HOT-DIP GALVANIZING LINE USING MACHINE LEARNING

S. S. Abdurakipov, E. B. Butakov
Kutateladze Institute of Thermophysics, Siberian Branch, Russian Academy of Sciences, Novosibirsk, Russia
Keywords: machine learning, gradient boosting, coverage defects, dross, digital assistant

Abstract

The paper presents a gradient boosting-based machine learning model developed to predict the occurrence of defects (dross) on galvanized steel sheets used in the automotive industry. An analysis of the influence of process parameters on the occurrence of defects is carried out, which makes it possible to identify the key factors affecting the quality of the coating: coil rolling speed, elongation and force in the coil-scin pass mill, zinc coating at top, temperature in the galvanizing pot, and temperature in the furnace snout. Based on the results of the study, a digital assistant is developed for evaluating coils in real time, providing simulation of the decision-making process and helping in prompt managing of the technological process.