The NVH performance of electric vehicles is a key indicator of vehicle quality, being the structure-borne transmission predominating at low frequencies. Many issues are typically generated by high vibrations, transmitted through different paths, and then radiated acoustically into the cabin. A combined analysis, with both finite-element and multi-body models, enables to predict the interior vehicle noise and vibration earlier in the development phases, to reduce the development time and moreover to optimize components with an increased efficiency level.
In the present work, a simulation of a Hyundai electric vehicle has been performed in IDIADA VPG with a full vehicle Multi-Body (MBD) model, integrated by vibration calculations with a Finite Elements (FE) model in MSC. Nastran to analyze the comfort.
Firstly, a full vehicle MBD model has been developed in MSC. ADAMS-Car including representative flexible bodies (generated from FE part models). The usage of a physical tire model and proper suspensions modelling has allowed to cascade time series loads induced by road interaction into the body.
A sequential approach (gradual validation: Component → Subsystem → Full vehicle) has been followed to validate the model up to 100 Hz for full vehicle modal response (for modal identification, both body and powertrain modes) and stationary maneuvers (constant velocity events).
The final part of the project has been dedicated to achieve proper correlation levels of the models in a global validation scenario, using different deterministic and random road surfaces, to identify appropriate vibrational levels as well as a good NVH predictivity.