Methods for fatigue under Random Vibration excitation for components of an Electric Vehicle

A vehicle travels on roads with various profiles of surface roughness that are best characterized by a random process. The random vibration not only subjects passengers to the unpleasant physical shakes and noises perceived, but also imparts the fatigue damage to components installed on it. The ability to predict fatigue damage at the early design stage to prevent unwanted durability problems down the road is very important in today’s design process, especially for an electronic vehicle. Although the random vibration analysis for a dynamic response for a linear system subjected to random excitations is fairly straightforward in the frequency-domain, the prediction of the fatigue damage as a result of those random responses is still an arena drawing a lot of research efforts. Different from the conventional approach to deal with the damage based on the stress or strain responses in a time-domain, the most effective method of predicting the fatigue in random vibration is to obtain the solution to the damage right in the frequency domain. In this work, the authors will apply this method to a very essential component in any current electric vehicle - battery pack housing - to illustrate how efficient it is in dealing with the fatigue prediction in this area. Various algorithms based on the key assumption for probability density functions for the stress range in the random response are compared to see how they may play a role in the accuracy of the fatigue prediction. As an interesting note, the authors will also show the importance of such a parameter as a modal damping ratio can dramatically affect the outcome of fatigue damage.