Improving the accuracy of time-harmonic FE simulations in induction heating applications: Towards better ferromagnetic material models. Issue 2 (6th March 2017)

• Record Type: Journal Article
• Authors: McMeekin, Kevin; Sirois, Frédéric; Tousignant, Maxime; Bocher, Philippe
• Is Part Of: Compel. Volume 36:Issue 2(2017)
• Page Start: 526
• Page End: 534
• Publication Date: 2017-03-06
• Subjects: Induction heating -- Finite element modeling -- Magnetic hysteresis -- Ferromagnetic materials -- Magnetic measurement; Electrical engineering -- Data Processing -- Periodicals; Electrical engineering -- Mathematics -- Periodicals; Electrical engineering -- Periodicals; Electronics -- Data Processing -- Periodicals; Electronics -- Mathematics -- Periodicals; 621.3
• DOI: 10.1108/COMPEL-05-2016-0203
• Languages: English
• ISSNs: 0332-1649

Abstract

Abstract : Purpose: Surface heat treatment by induction heating (10-100 kHz) requires precise prediction and control of the depth of the induced phase transformation. This paper aims at identifying common issues with the measurement and modeling of magnetic properties used in induction heating simulations, and it proposes ways to improve the situation. Design/methodology/approach: In particular, it is demonstrated how intrinsic magnetic properties (i.e. the B-H curve) of a sample can change during the magnetic characterization process itself, due to involuntary annealing of the sample. Then, for a B-H curve that is supposed perfectly known, a comparison is performed between multiple models, each one representing the magnetic properties of steel in time-harmonic (TH) finite element method simulations. Finally, a new model called "power-equivalent model" is proposed. This model provides the best possible accuracy for a known nonlinear and hysteretic B-H curve used in TH simulations. Findings: By carefully following the guidelines identified in this paper, reduction of errors in the range of 5-10 per cent can be achieved, both at the experimental and modeling levels. The new "power-equivalent model" proposed is also expected to be more generic than existing models. Originality/value: This paper highlights common pitfalls in the measurement and modeling of magnetic properties, and suggests ways to improve the situation.