Improvement of an SEA Model of Cab Interior Sound Levels Through Use of a Hybrid FE/SEA Method

An existing system-level Statistical Energy Analysis (SEA) model of an enclosed operator station (cab) for a combine-harvester was improved through component-level analyses using Finite Element (FE) and hybrid FE-SEA methods. At mid to high audio frequencies, airborne transmission of machine noise is a dominant path for the cab. An SEA model was created for the cab using the VA One product. When model results were validated against experimental data derived from three idealized insonification load cases, the original model did not compare well with the measured data. The structural panels used in the cab feature various non-uniform cross-sections and varying radii of curvature. The former are not appropriately modeled with standard beam stiffeners, and the latter must be accounted for by some average curvature. Geometrically accurate Finite Element (FE) models of the panels were employed to estimate parameters including effective material stiffness, and effective material density. Further, hybrid FE-SEA modeling was employed to estimate panel radiation efficiencies and account for non-uniform curvature. Using the corrected parameters, the improved SEA model compared well with measured data; it can be used with confidence to predict cab interior sound levels for different field load cases, and to explore noise control treatments.