Fatigue Life Prediction on Rough Road Using Full Vehicle Co-simulation Model with Suspension Control

<div>The ability to predict the durability of a structure depends on the knowledge of operating loads experienced by the structure. Typically, multi-body dynamics (MBD) models are used to cascade measured wheel loads to hard points. However, in this approach, there are many sources by which errors creep into cascaded forces. Any attempt to reduce sources of such errors is time consuming and costly. In typical program development timelines, it is very difficult to accommodate such model calibration efforts. Commercial load cells exist in the industry to give engineers insight into understanding the complex real-world loading of their structures. A significant limitation to the use of load cells is that the structure needs to be modified to accept the load cell, and not all desired loading degrees of freedom (DOFs) can be measured. One of the innovative solutions to calculate operating loads is to convert the structure itself into its own load transducer. The D-optimal algorithm along with the pseudo-inverse technique provides a theoretically sound and versatile method to identify optimum positions and locations to place the sensors (i.e., strain gauges) on the structure where its response is to be measured. A pre-calculated calibration matrix through pseudo-inverse is then used along with measured responses to reverse calculate loads acting on the structure. The accuracy of calculated loads with this approach is typically high compared with conventional load cascading methods as sources of errors are less in this method.</div> <div>This work is focused on load reconstruction, FE analysis, and lightweighting of the bell crank lever of a commercial vehicle. Practical difficulties associated with the load reconstruction method and solutions are also discussed in this research paper.</div>

show abstract