Prediction of vibration environment in real operations described by Q-distributions

The purpose of this paper is to present a method which makes it possible to predict the vibration environment of a vehicle in real operations. By using the Q-distribution, which defines the distribution of the operating severity, the non-stationary nature of the road-excitation can be described. The procedure is exemplified by predicting the fatigue life of a heavy vehicle leaf spring. It is not possible to predict the fatigue life if stationary random input is assumed. This leads to a substantial over-estimation. It is sufficient to use mathematical models with a low degree of complexity and few degrees of freedom to obtain useful predictions. Non-linear descriptions of dampers and bump-stops in the model are necessary in order to achieve quantitatively good estimates. The Q-distribution can be used to identify the potential to differentiate and optimize the design for different operating conditions and vehicle applications.