Modal Overlap at Low Frequencies - A Stochastic Approach for Vehicle System Modal Management 2003-01-1612

In the early stages of a vehicle program, it is a common practice to set target ranges for the global body, suspension and powertrain modes. This modal management process allows engineers to avoid potential noise and vibration problems stemming from strong overlap of major global modes. Before the first prototype hardware is built, finite element models of the body, suspension and powertrain are usually exercised to compare predicted versus targeted ranges of the major system modes in the form of a modal management chart. However, uncertainty associated with the design parameters, manufacturing process and other sources can lead to a major departure from the design intent when the first hardware prototype is built. In this study, a first order reliability method is used to predict variance of the eigen values due to parameter uncertainties. This allows the CAE engineers to add a “three sigma” bound on the eigen values reported in the modal management chart. In addition, a modified form of the Modal Overlap Factor (MOF) borrowed from Statistical Energy Analysis (SEA) is employed to develop a method of assessing the “risk” of modal overlap. The theoretical basis for the method is first established followed by an example involving a discrete dynamical system; a 4-dof “half-car” analytical model. Finally, applicability of the method is demonstrated on a production vehicle system FE model.