In automotive market, today more than in the past, it is very important to reduce time to market and, mostly, developing costs before the final production start. Ideally, bench and on-road tests can be replaced by multi-body studies because virtual approach guarantees test conditions very close to reality and it is able to exactly replicate the standard procedures. Therefore, today, it is essential to create very reliable models, able to forecast the vehicle behavior on every road condition (including uneven surfaces).
The aim of this study is to build an accurate multi-body model of a heavy-duty truck, check its handling performance, and correlate experimental and numerical data related to comfort tests for model tuning and validation purposes. Experimental results are recorded during tests carried out at different speeds and loading conditions on a Belgian blocks track. Simulation data are obtained reproducing the on-road test conditions in multi-body environment. The virtual vehicle is characterized by rigid and flexible bodies, the tire model used is FTire (Flexible Structure Tire Model) while the 3D scan of the road surface is imported using OpenCRG format. Signals coming from accelerometers, positioned on suspension axles, the truck chassis and cabin, are investigated both in time and frequency domain, using three different methods typical of random signals: the power spectral density (PSD) analysis, the root mean square (RMS) and the level crossing peak counting (LC). The good match between simulation and experimental data validates the adopted simulation methodology, therefore this work has provided a valuable tool for studies concerning comfort, NVH, durability and fatigue in order to improve safety and reliability of future heavy-duty vehicles.