Noise, Vibration and Harshness (NVH) of Electric Powertrains: Electric Motor Whistling vs Transmission Whining

While ICEs tend to be louder, they emit sound in a broadband manner usually below 1 kHz. E-powertrains are quieter, but radiate noise in distinct frequencies governed by the powertrain topology and motor speed, at frequencies that can be higher than 10 kHz.

Two key contributors are usually present: e-motor whistling and gear whining. The electromagnetic forces acting on the motor’s stator and rotor are the source of whistling. Gear whining occurs because of the time-varying meshing stiffness of the gears. The tonal behaviour from these two sources can have a detrimental effect on the perceived sound quality of EVs.

Our Aim

Our research aims to identify the electromechanical coupling in e-powertrains and its effect on e-motor whistling and transmission whining NVH. We develop novel design methods to reduce their severity and avoid costly remedial NVH measures later in the development process.

Our Research

We employ fully flexible 3D multi-body dynamics to capture the contributions from the electromagnetic forces and the gear meshing, along with local non-linearities and predict the surface vibration of the e-powertrain housing along with the radiated noise.

Artificial Neural Networks

In order to provide fast design solutions tackling the perception of the tonal content of radiated noise, we develop artificial neural networks (ANNs) to optimise the powertrain and assess psychoacoustic features based on bearing forces transmitted to the housing. We use the Prominence Ratio (PR) metric to assess if a frequency in the noise spectrum will be perceived as tonal or not.

The effects of electromechanical coupling

Our applied research also investigates the effects of the electromechanical coupling on the natural frequencies and mode shape of the powertrain, identifying potentially aggressive whistling or whining for practical design solutions.

Finally, we develop reduced-order models for NVH predictions in e-powertrains using 2-Dimensional (2D) multiphysics, delivering results within minutes for Permanent Magnet Synchronous Machines under load. The system's vibroacoustic response is computed by employing equivalent 2D structural models that represent the stator, and acoustic analyses is performed over the complete speed range of the e-motor.

By utilizing numerical and analytical expressions, our methods provide direct insight into the NVH generation mechanisms of e-powertrains, often slicing the simulation time that is required by complex finite-element based methodologies. These benefits are paramount for NVH engineers, leading to quieter and more refined e-powertrains within shorter development times.

Our research is funded by the Engineering and Physical Sciences Research Council (EP/T518098/1 – DTP 2020-2021 含羞草视频) & (EP/V053353/1 – Automotive electric powertrain whistling and whining: fundamental root cause analysis to novel solutions) and UK Research and Innovation (2585407 – Electric motor and transmission coupled vibro-acoustics of electric powertrains).

Research dissemination

Our research has been promoted and disseminated at International Conferences, such as Aachen Acoustics Colloquium, ISNVH, NAFEMS, SAE World Congress and via social media channels.

Research members

• Dr Amal Hajjaj
• Dr Mahdi Mohammad-Pour
• Prof Stephanos Theodossiades

Selected publications

1. Andreou, P, Theodossiades, S, Hajjaj, AZ, Mohammadpour, M, Ricardo Souza, M (2024) Reduced Order Model for Modal Analysis of Electric Motors Considering Material and Dimensional Variations. SAE Technical Papers. DOI: 10.4271/2024-01-2945.
2. Ricardo Souza, M, Offner, G, Mohammadpour, M, Andreou, P, Theodossiades, S (2024) Metrics Based Design of Electromechanical Coupled Reduced Order Model of an Electric Powertrain for NVH Assessment. SAE Technical Papers. DOI: 10.4271/2024-01-2913.
3. Morris, N, Andreou, P, Ricardo Souza, M, Mohammad pour, M, Theodossiades, S (2024) Tribodynamic analysis of electric vehicle powertrains. In Electric Vehicle Tribology Challenges and Opportunities for a Sustainable Transportation Future, Elsevier, pp.41-56, ISBN: 9780443140754.
4. Souza, MR, Haris, A, Rodrigues, L, Offner, G, Sopouch, M, Diwoky, F, Mohammad-Pour, M, Theodossiades, S (2024) The use of an artificial neural network for assessing tone perception in electric powertrain noise, vibration and harshness, Meccanica, 59(3), pp.433-459, ISSN: 0025-6455. DOI: 10.1007/s11012-024-01753-x.
5. Andreou, P, Hajjaj, A, Mohammad-Pour, M, Theodossiades, S (2023) Analytical multiphysics methodology to predict vibroacoustics in PMSMs combining tangential electromagnetic excitation and tooth modulation effects, IEEE Transactions on Transportation Electrification, ISSN: 2332-7782. DOI: 10.1109/TTE.2023.3325350.
6. Theodossiades, S, Morris, N, Mohammadpour, M (2022) On the Road Towards Zero-Prototype Development of Electrified Powertrains via Modelling NVH and Mechanical Efficiency. In Advances in Engine and Powertrain Research and Technology, Springer International Publishing, pp.267-290, ISBN: 9783030918682. DOI: 10.1007/978-3-030-91869-9_11.
7. Souza, M.R., Mohammadpour, M., Theodossiades, S., Diwoky, F. et al., “NVH of Electric Motors - A Study on Potential NVH Metric,” in Aachen Acoustics Colloquium Proceedings, Aachen, 2022.
8. Andreou, P, Zemni, AHE, Mohammad-Pour, M, Theodossiades, S (2022) Analytical multiphysics model for NVH prediction of a high-speed Surface-Permanent Magnet Synchronous Machine. In 53rd Spanish Congress of Acoustics 2022; Proceedings of the Spanish Congress of Acoustics, Alicante, Spain.