Acoustic noise computation of electrical motors using the boundary element method

Sabin Sathyan*, Ugur Aydin, Anouar Belahcen

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

17 Citations (Scopus)
116 Downloads (Pure)


This paper presents a numerical method and computational results for acoustic noise of electromagnetic origin generated by an induction motor. The computation of noise incorporates three levels of numerical calculation steps, combining both the finite element method and boundary element method. The role of magnetic forces in the production of acoustic noise is established in the paper by showing the magneto-mechanical and vibro-acoustic pathway of energy. The conversion of electrical energy into acoustic energy in an electrical motor through electromagnetic, mechanical, or acoustic platforms is illustrated through numerical computations of magnetic forces, mechanical deformation, and acoustic noise. The magnetic forces were computed through 2D electromagnetic finite element simulation, and the deformation of the stator due to these forces was calculated using 3D structural finite element simulation. Finally, boundary element-based computation was employed to calculate the sound pressure and sound power level in decibels. The use of the boundary element method instead of the finite element method in acoustic computation reduces the computational cost because, unlike finite element analysis, the boundary element approach does not require heavy meshing to model the air surrounding the motor.

Original languageEnglish
Article number245
Number of pages13
Issue number1
Publication statusPublished - 3 Jan 2020
MoE publication typeA1 Journal article-refereed


  • Acoustics
  • Boundary element method
  • Electric machines
  • Finite element method
  • Induction motors
  • Magneto-mechanics
  • Modeling
  • Noise
  • Vibro-acoustics


Dive into the research topics of 'Acoustic noise computation of electrical motors using the boundary element method'. Together they form a unique fingerprint.

Cite this