A Universally Applicable Methodology for Comparing, Calibrating, and Validating the Performance of Multi-Wheeled Combat Vehicle Computer Simulation Models to Experimental Test Data 2006-01-1164

The dynamic performance of a multi-wheeled combat vehicle developed in a multi-body dynamics code initially on the basis of design information was compared, calibrated and validated in a systematic three-stage process against measured data obtained on US Army Aberdeen Test Center (ATC) test courses. In the comparison stage the model was refined using insights gained from the live vehicle tests. At that point, the model was adjusted or calibrated using optimization software. The process of optimization required judicious choice of performance parameters suited to the particular test event referencing a test course or maneuver. As an example, the vehicle roll, pitch and yaw rates were used to characterize the vehicle for the J-turns, whereas statistical moments of the vertical acceleration were selected to describe the vehicle performance on the 6-inch Washboard. Once optimized the model was subjected to a further limited validation using experimental live vehicle test data held in abeyance during the comparison phase. Comparisons were carried out in both the time and frequency domains, via rms, skewness, and kurtosis and power spectral densities as well as shock response spectrums respectively. An equation representing a “figure of merit” was defined and computed for each of the test courses and maneuvers. These quantities were combined in an overall weighted sum to describe the general performance of the model across all test events. The methodology used to develop the model is expected to serve as a measure of the utility of a multi-body dynamics model in the development dynamics model in the development phase of a vehicle design, prior to the availability of a test item. Longer term the model is expected to be of considerable benefit as a reliability tool by which accurate force and acceleration data derived from the model can be applied to predict the fatigue life of components in the field.