The Optimization of Distribution and Fan Parameters in Heat Treatment Furnaces Through the Integration of Numerical Simulation and Machine Learning

Abstract

The present study employed numerical simulation technology to investigate the distribution of workpieces within a low-temperature trolley heat furnace and analyze the influence of circulating fan parameters on heat treatment quality. This analysis was integrated with machine learning technology to guide heat treatment production. The research findings indicate that when the number of workpieces remains constant, their position has a significant impact on airflow velocity distribution, heating rate, and temperature uniformity within the furnace. Additionally, wind pressure from the circulating fan affects both fluid field and temperature field; the increasing wind pressure leads to higher flow rates in the furnace as well as increases heating rates for workpieces. Heating efficiency exhibits a nonlinear relationship with wind pressure increment. By adjusting air pressure distribution from the circulating fan, workpiece temperature uniformity can be improved by 64%. Furthermore, machine learning technique demonstrates excellent performance in predicting workpiece temperatures with a maximum relative error of 2.4%, while maintaining consistent trends in temperature uniformity.