Manuel Islam, M.Sc.

Industrial PhD

Methodology for evaluating the vibroacoustic characteristics of automotive electric drives considering multiphysics effects

The automotive sector is undergoing a drastic transformation in which electrified powertrains are increasingly replacing conventional combustion engines. Accordingly, the requirements and challenges in developing electric drivetrains are changing. The acoustic characteristics of an electric drivetrain, particularly the electric motor, represent a significant challenge as the broadband acoustic masking provided by the internal combustion engine no longer applies. Furthermore, the operation of such an electric machine falls into a frequency range that is particularly sensitive to the human ear and is generally perceived as unpleasant as well as annoying.

In addition to the rotor, the stator of an electric radial flux machine in particular contributes to noise emission. A fundamental cause of these noises and vibrations is the inherent electromagnetic force of the machine. In this regard, perceptible vibrations and, thus, acoustic noises occur when the frequency and the mode of the exciting electromagnetic force match the natural frequency and the mode of vibration of the stator. Consequently, knowing the parameters that influence the natural frequency and the mode shapes of the stator is an essential foundation in the development process. Conducting an experimental modal analysis (EMA) is crucial for gaining such knowledge of structural dynamic parameters. Subsequently, these results and findings are incorporated into the numerical analysis. Thus, modeling methodologies are elaborated and the prediction quality of numerical models is improved. Because of the limited research activities that consider the influence of temperature, it remains to be seen how an unwound stator and hence the lamination behave structure-dynamically at elevated temperatures. Also, it is debatable whether the additional stator winding affects the structural dynamic behavior under higher temperatures.

Therefore, there is a particular need to develop a modeling methodology for lamination and winding that can represent temperature effects. The aim is to consider these influences and provide a modeling methodology to improve prediction quality in the virtual prototyping of electrical machines.

Research Topics

  • Finite Element Method
  • Parameter identification
  • Multiphysics problems (electromagnetism, structural dynamics, heat transfer)

Project Partners

  • Technical University of Munich
  • Mercedes-Benz AG