A COMPUTER-AIDED FINITE-ELEMENT EXPERIMENTAL-ANALYSIS OF INDUCTION-HEATING PROCESS OF STEEL

Citation
K. Sadeghipour et al., A COMPUTER-AIDED FINITE-ELEMENT EXPERIMENTAL-ANALYSIS OF INDUCTION-HEATING PROCESS OF STEEL, Computers in industry, 28(3), 1996, pp. 195-205
Citations number
8
Categorie Soggetti
Computer Application, Chemistry & Engineering","Computer Science Interdisciplinary Applications
Journal title
ISSN journal
01663615
Volume
28
Issue
3
Year of publication
1996
Pages
195 - 205
Database
ISI
SICI code
0166-3615(1996)28:3<195:ACFEOI>2.0.ZU;2-P
Abstract
The characteristics of steels during electroplating or induction heati ng are very complex. Induction heat-treating of steel has become incre asingly important as a means of reducing cost and improving the qualit y of carbon steel components. The use of high frequency (40 to 200 kHz ) offers many advantages, but tooling and cycle optimization is comple x and time consuming. Approximate analytical methods suitable for low- frequency approximations are not sufficient for the low-frequency doma in and the fact that important material properties change drastically with temperature makes more exact analysis methods very difficult to i mplement. Therefore, a powerful computer aided numerical tool (i.e., f inite element analysis) is selected to numerically model an induction heating process. A general-purpose finite element program was employed to simulate and analyze the above problem. The combination of magneti c and thermal routines, within the package, enabled us to complete the task. A coupling method between the two magnetic and thermal routines was also developed and implemented. This was done to incorporate the change in the material properties due to the change in temperature. Fo r the sake of comparison and verification, a high-frequency induction heating experiment was set up and a series of tests was performed. The finite element results were evaluated and compared with the experimen tal results. The effect of the time duration size (time step) for coup ling between the magnetic and thermal analyses was also studied. Parti cular attention was given to the coupling procedure after the Curie te mperature was reached. The skin effect was studied and demonstrated in the numerical model. A discussion of induced power density profile an d transient temperature distribution is also presented. The accuracy a nd efficiency of the numerical models are demonstrated and appreciated . This tool is therefore proposed to be a powerful alternative prior t o the actual induction heating process.