Analysis of FE: Mesh discretization for faster and more accurate probabilistic simulations: applied to cogging torque calculations in electrical machines. Issue 2 (2nd March 2015)

• Record Type: Journal Article
• Authors: Offermann, Peter; Hameyer, Kay
• Editors: Slawomir Wiak, Professor
• Is Part Of: Compel. Volume 34:Issue 2(2015)
• Page Start: 596
• Page End: 603
• Publication Date: 2015-03-02
• Subjects: Finite element method -- Cogging torque -- Monte-Carlo simulation -- Uncertainty quantification; 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-07-2014-0174
• Languages: English
• ISSNs: 0332-1649

Abstract

Abstract : Purpose: – The consideration of uncertainties in the numerical computation of electromagnetic fields has recently gained a lot of attention. Most publications focus on the creation of models for the uncertainty quantification, however, neglect the inaccuracy of the applied finite element model itself. Thus, the purpose of this paper is to analyze the influence of mesh quality on stochastic cogging torque calculations. Design/methodology/approach: – The presented work consists of three steps. At first, a conventional analysis of the influence of mesh accuracy onto cogging torque is presented. Afterwards, the method is extended to stochastic calculations. Based on a comparison of the convergence behavior of both approaches, a method for more accurate cogging torque predictions with fewer necessary calculations is derived. Findings: – An improved method to calculate probability predictions at minimum computational cost is presented and applied. Research limitations/implications: – The presented approach requires the exact knowledge of the system's stochastic variation boundaries. Originality/value: – A fast method for more accurate stochastic cogging torque calculations is developed.