Dipl.-Ing. Dimitrios Ernst Tsokaktsidis

Industrial PhD

Machine Learning and NVH-transfer in the automotive sector

Vehicle development processes are increasingly digitalized in order to satisfy growing market needs and economic factors. Depending on the frequency range of interest, different simulation methods (the Multibody Simulation, the Finite Element Method, the Boundary Element Method and the Statistical Energy Analysis) have been established within the NVH-departments. The spectral limitations, the manual system modelling, the accuracy of the calculation results and the calculation duration offer opportunities for improvement. The possible application of machine learning in the described context is investigated.

Research Topics

  • Supervised learning
  • Unsupervised learning
  • Transfer path analysis

Project Partners

  • Technical University of Munich
  • Mercedes-Benz AG



  • Tsokaktsidis, Dimitrios Ernst; Nau, Clemens; Maeder, Marcus; Marburg, Steffen: Using rectified linear unit and swish based artificial neural networks to describe noise transfer in a full vehicle context. The Journal of the Acoustical Society of America 150 (3), 2021, 2088-2105 more… BibTeX Full text ( DOI )


  • Tsokaktsidis, Dimitrios Ernst; Nau, Clemens; Marburg, Steffen: Time Domain Full Vehicle Interior Noise Calculation from Component Level Data by Machine Learning. SAE Technical Paper Series, SAE International, 2020 more… BibTeX Full text ( DOI )


  • Tsokaktsidis, Dimitrios Ernst; Wysocki, Timo Von; Gauterin, Frank; Marburg, Steffen: Artificial Neural Network predicts noise transfer as a function of excitation and geometry. Proceedings of the ICA 2019 and EAA Euroregio : 23rd International Congress on Acoustics integrating 4th EAA Euroregio 2019 : 9-13 September 2019, 2019, 9 Sep 2019-13 Sep 2019; Aachen (2019). more… BibTeX Full text ( DOI )