Natural Frequency Analysis of Modelica Powertrain Models

This document explores the development of automated natural frequency and modal analysis methods for complex vehicle powertrain models created in Dymola using Modelica and the PTDynamics library.

The research focuses on overcoming challenges associated with analysing highly detailed multi-body powertrain systems, including relative state selection, friction component behaviour, clutch dynamics, and tyre slip modelling. By introducing automated state transformation and linearisation techniques, the study enables accurate prediction of resonance frequencies, modal responses, and NVH (Noise, Vibration, and Harshness) characteristics for drivetrain systems under different operating conditions. Using both simplified and full vehicle powertrain models, the research demonstrates how advanced modal analysis tools can improve understanding of driveline dynamics, support drivability optimisation, and enhance vehicle powertrain design and refinement processes.