A Methodology for Efficient Conceptual Design Analysis of Nonlinear Dynamic Structural Systems

Licentiate thesis, 2013

There is a typical requirement conflict between fuel consumption and noise and vibrations in passenger cars. Obviously the combustion engine is a major source of vibration and has an influence on vehicle emissions. Similar properties apply to the power transferring mechanical driveline system. Many driveline design concepts that aim to reduce carbon-dioxide emissions intensify and add new vibration problems, since the majority of them affect vibration sources or system damping. In conceptual design of drivelines, many possible concepts proposals are studied in parallel. This situation calls for modelling and analysis that can meet the demand for rapid virtual prototyping. Conflicting to this is the trend in which models have become extremely detailed to meet demands from others than conceptual designers. The complex behaviour of the driveline system in combination with often highly specialised component models result in system models that at their best are valid only near a few specific stationary operating points. This makes is difficult to study the effectiveness of possible component design changes in early development phases. Instead, this is first revealed during system verification testing, when fundamental design changes since long are unrealisable. This work focus on models for conceptual design, \emph{i.e.} models that are not overparameterized. The aim is to find the balance when models are as simple as possible but as accurate as required by conceptual design studies. A methodology is proposed, based on knowledge of existing automotive methods and workflow, that with provided modelling tools has the potential to serve this purpose within the nearest future.